1 00:00:00,000 --> 00:00:14,520 Alright, so let's begin. 2 00:00:14,520 --> 00:00:17,540 So just a little bit of background here, Alice. 3 00:00:17,540 --> 00:00:20,220 So I'll put in the chat. 4 00:00:20,220 --> 00:00:27,340 So we've been working with some of the Washington State data. 5 00:00:27,340 --> 00:00:36,720 So in specific, you can make public records requests and actually get the traceability 6 00:00:36,720 --> 00:00:38,400 data dumps. 7 00:00:38,400 --> 00:00:52,000 So for example, let me find the specific link and I will put it in the chat. 8 00:00:52,000 --> 00:01:05,400 So here you can access, that would be a snapshot of what the data would look like in December. 9 00:01:05,400 --> 00:01:10,400 So when was your group created and what's the goal of the group? 10 00:01:10,400 --> 00:01:17,640 So the group is just to sort of share what we're working on. 11 00:01:17,640 --> 00:01:23,920 So all of us are sort of working tangentially with the cannabis industry. 12 00:01:23,920 --> 00:01:29,880 And so we're trying to do some analysis to just get a better understanding of the industry 13 00:01:29,880 --> 00:01:32,160 and of particular markets. 14 00:01:32,160 --> 00:01:40,460 So it's a group to bring together like-minded people that want to talk about cannabis data. 15 00:01:40,460 --> 00:01:44,000 And so it's actually a pretty open-ended group. 16 00:01:44,000 --> 00:01:48,640 So at the moment, I've just been sharing some of the work I've been doing. 17 00:01:48,640 --> 00:01:56,000 Charles has been doing a lot of data cleaning on some of the data we've been able to come 18 00:01:56,000 --> 00:01:57,400 across. 19 00:01:57,400 --> 00:02:03,040 And then some people have come in that were interested in APIs. 20 00:02:03,040 --> 00:02:10,800 And today I was going to share with you some data that I was looking at in Colorado a few 21 00:02:10,800 --> 00:02:16,520 years ago and see if anyone's interested in that data. 22 00:02:16,520 --> 00:02:21,920 And if so, then it may be worthwhile updating it with the most recent data. 23 00:02:21,920 --> 00:02:25,200 And are you both involved in the cannabis industry? 24 00:02:25,200 --> 00:02:26,200 Yes. 25 00:02:26,200 --> 00:02:35,000 Well, I can't speak for Charles, but I've started a company, Cannelytics, that provides 26 00:02:35,000 --> 00:02:36,480 cannabis analytics. 27 00:02:36,480 --> 00:02:40,920 Primarily, we focus on providing solutions to laboratories. 28 00:02:40,920 --> 00:02:47,400 So we're providing an end-to-end solution for laboratories that need a laboratory information 29 00:02:47,400 --> 00:02:48,680 management system. 30 00:02:48,680 --> 00:02:55,720 And we also provide cannabis analytics to producers, retailers, and consumers. 31 00:02:55,720 --> 00:03:03,720 And then just part-time, I do some data scientist work for a company in San Francisco that does 32 00:03:03,720 --> 00:03:06,840 supply chain management in the cannabis industry. 33 00:03:06,840 --> 00:03:07,840 OK. 34 00:03:07,840 --> 00:03:11,120 And what about you, Charles? 35 00:03:11,120 --> 00:03:13,000 I'm just interested in data science. 36 00:03:13,000 --> 00:03:17,280 And this is a unique data set that not a lot of people have studied. 37 00:03:17,280 --> 00:03:23,360 And so I'm interested in looking at it and seeing what I can find. 38 00:03:23,360 --> 00:03:24,360 Awesome. 39 00:03:24,360 --> 00:03:29,840 Well, speaking of which, let me go ahead and... 40 00:03:29,840 --> 00:03:50,640 Let me share with you some more resources. 41 00:03:50,640 --> 00:03:58,600 So I guess I'll just go ahead and start sharing my screen, and we can kick it off for today. 42 00:03:58,600 --> 00:04:01,680 All right. 43 00:04:01,680 --> 00:04:07,080 All right. 44 00:04:07,080 --> 00:04:20,120 So some of the data sources that I've been looking at were... 45 00:04:20,120 --> 00:04:31,640 A while back, I had started writing just this paper, and I was looking at wages. 46 00:04:31,640 --> 00:04:48,560 And I recently saw this post by Leafly. 47 00:04:48,560 --> 00:05:14,320 And I'll also put this link in the chat. 48 00:05:14,320 --> 00:05:18,480 So long story short, it's a post by Leafly. 49 00:05:18,480 --> 00:05:27,160 And they're talking about how there's a large number of jobs in the cannabis industry. 50 00:05:27,160 --> 00:05:29,320 It's a growing field. 51 00:05:29,320 --> 00:05:35,120 And that was an area that I'd looked at previously. 52 00:05:35,120 --> 00:05:41,560 So I thought it would be worthwhile to do some analysis. 53 00:05:41,560 --> 00:05:49,720 So the first place we need is just a nice source of data. 54 00:05:49,720 --> 00:05:58,400 And for no particular reason, but I just chose Colorado. 55 00:05:58,400 --> 00:06:04,600 Their cannabis industry is fairly well established, and they have a fairly good source of public 56 00:06:04,600 --> 00:06:06,160 data. 57 00:06:06,160 --> 00:06:19,280 So I'll put these links in the chat here, but there are some... 58 00:06:19,280 --> 00:06:34,920 There's some good data about the actual number of licensees in Colorado. 59 00:06:34,920 --> 00:06:37,840 It's been legal there for quite a while. 60 00:06:37,840 --> 00:06:38,840 Right? 61 00:06:38,840 --> 00:06:39,840 Exactly. 62 00:06:39,840 --> 00:06:42,480 So they've had medicinal... 63 00:06:42,480 --> 00:06:46,600 I don't want to say how long, but probably more than 10 years. 64 00:06:46,600 --> 00:06:56,080 And then the recreational dates back to really the start of 2015. 65 00:06:56,080 --> 00:06:58,160 So there's more data there. 66 00:06:58,160 --> 00:06:59,160 Exactly. 67 00:06:59,160 --> 00:07:14,120 So just taking a quick look at some of this data, you can actually go to the API docs. 68 00:07:14,120 --> 00:07:24,520 And so this is useful because you can interact with this data in your favorite programming 69 00:07:24,520 --> 00:07:25,520 language. 70 00:07:25,520 --> 00:07:31,440 So they even have all these code snippets down here at the bottom. 71 00:07:31,440 --> 00:07:33,720 I just tend to use Python. 72 00:07:33,720 --> 00:07:41,160 So I'm just going to just show you just a quick example of how to hit this endpoint 73 00:07:41,160 --> 00:07:46,160 in Spyder just to show you how to get some licensee data. 74 00:07:46,160 --> 00:07:49,760 So it's fairly simple. 75 00:07:49,760 --> 00:07:59,960 I'll be using Socrata's development tools. 76 00:07:59,960 --> 00:08:04,560 And then I'm just going to use an unauthenticated client. 77 00:08:04,560 --> 00:08:11,340 However, you can get an API key just for authentication. 78 00:08:11,340 --> 00:08:22,760 Just for authentication, that way you don't have to be throttled. 79 00:08:22,760 --> 00:08:33,720 And then the way you really work with the endpoints is there's a unique ID for a given 80 00:08:33,720 --> 00:08:35,400 data set. 81 00:08:35,400 --> 00:08:43,360 So for example, when we're looking here at this data set, you'll see this ID right there 82 00:08:43,360 --> 00:08:47,520 at the end of the URL. 83 00:08:47,520 --> 00:08:56,680 And then likewise, the same ID. 84 00:08:56,680 --> 00:09:01,760 And then you can just use standard requests. 85 00:09:01,760 --> 00:09:09,440 But in this case, I'll just use the client, and I'm just going to get maybe just 10 observations 86 00:09:09,440 --> 00:09:10,440 here. 87 00:09:10,440 --> 00:09:15,080 So you can make a request. 88 00:09:15,080 --> 00:09:25,520 I'm not exactly sure what type of response object this is. 89 00:09:25,520 --> 00:09:41,160 But we can convert it to a Pandas data frame. 90 00:09:41,160 --> 00:09:58,080 And so then you can look at observations. 91 00:09:58,080 --> 00:10:04,680 So I didn't actually save this data to the GitHub repository, because it does have people's 92 00:10:04,680 --> 00:10:05,880 actual names. 93 00:10:05,880 --> 00:10:18,760 So I guess that's public information, but it's of no interest to me. 94 00:10:18,760 --> 00:10:31,240 I essentially just went and did a count of the number of licensees by county. 95 00:10:31,240 --> 00:10:33,600 I think you get people's city. 96 00:10:33,600 --> 00:10:37,560 So from the city, you can deduce the county. 97 00:10:37,560 --> 00:10:45,360 And then you can just get a nice county count. 98 00:10:45,360 --> 00:10:48,200 There's other fields here that may be of interest. 99 00:10:48,200 --> 00:10:52,840 Not many people fill them out, but there are degrees. 100 00:10:52,840 --> 00:11:12,840 So some people will list their degrees and their 101 00:11:12,840 --> 00:11:14,500 specialties. 102 00:11:14,500 --> 00:11:34,960 So that could be an interesting analysis to look at what specialties and degrees people 103 00:11:34,960 --> 00:11:39,920 in the cannabis industry tend to fall into. 104 00:11:39,920 --> 00:11:48,200 Because there's people from all walks of life in the cannabis industry. 105 00:11:48,200 --> 00:11:56,480 So just to get back on track here, just to show you the actual work I was doing here. 106 00:11:56,480 --> 00:12:07,280 So I'll actually show you the data. 107 00:12:07,280 --> 00:12:10,560 So I did recently... 108 00:12:10,560 --> 00:12:24,240 Sorry, can I ask you in this data set, so professional and occupational license, are 109 00:12:24,240 --> 00:12:28,080 we talking about people who work for a certain business or are we talking about the businesses 110 00:12:28,080 --> 00:12:29,080 themselves? 111 00:12:29,080 --> 00:12:32,680 I'm looking at the data from right now. 112 00:12:32,680 --> 00:12:40,080 So in Colorado, you have to have a license to work in the cannabis industry. 113 00:12:40,080 --> 00:12:43,720 So these are employee specific licenses. 114 00:12:43,720 --> 00:12:49,600 So I think they're tied to a specific person. 115 00:12:49,600 --> 00:12:56,040 So yeah, is there somewhere where it says the company that they work in, if they are 116 00:12:56,040 --> 00:12:57,400 working in the company? 117 00:12:57,400 --> 00:13:03,040 I was looking at that and I couldn't find that because that would be... 118 00:13:03,040 --> 00:13:18,280 Like you kind of hit the nail on the head, that would be an incredibly interesting breakdown 119 00:13:18,280 --> 00:13:22,320 to look at firm size. 120 00:13:22,320 --> 00:13:23,320 Right? 121 00:13:23,320 --> 00:13:24,320 Yeah. 122 00:13:24,320 --> 00:13:30,240 I mean, in the leafy graph that you had, the infographics, it was... 123 00:13:30,240 --> 00:13:33,480 Or maybe in one of your slides, you say you have... 124 00:13:33,480 --> 00:13:36,680 The company has between one and 10 employees or... 125 00:13:36,680 --> 00:13:37,920 Oh, yes. 126 00:13:37,920 --> 00:13:43,160 So now I was going to kind of show you some of the... 127 00:13:43,160 --> 00:13:46,760 This is some of the research that I was doing. 128 00:13:46,760 --> 00:13:52,920 So we're here talking about data science applied to the cannabis industry. 129 00:13:52,920 --> 00:14:03,560 And so the way you can really capitalize and make interesting observations is to combine 130 00:14:03,560 --> 00:14:05,360 data sets. 131 00:14:05,360 --> 00:14:12,280 So you can get pretty rich data out of one data set. 132 00:14:12,280 --> 00:14:18,600 But now if we start to combine data sets, we may have to make some approximations and 133 00:14:18,600 --> 00:14:25,160 estimations, but we can start to get some insights here. 134 00:14:25,160 --> 00:14:35,440 So this was a survey that was done in 2014. 135 00:14:35,440 --> 00:14:37,720 So this is going to be dated. 136 00:14:37,720 --> 00:14:43,200 And plus, I did this research in 2017. 137 00:14:43,200 --> 00:14:50,160 So the things I'm sharing with you are almost four years old. 138 00:14:50,160 --> 00:14:52,240 And so it would be worthwhile... 139 00:14:52,240 --> 00:14:59,040 And that's why I'm kind of working on them is because these are things I did four years 140 00:14:59,040 --> 00:15:06,040 ago and it may be worthwhile to touch them up with current data. 141 00:15:06,040 --> 00:15:11,120 So that's why I'm sharing it with you in case any of you want to work on this type of stuff 142 00:15:11,120 --> 00:15:12,120 as well. 143 00:15:12,120 --> 00:15:23,680 But long story short here, I found this work and wellbeing survey and I'll have to share 144 00:15:23,680 --> 00:15:27,880 this link after the presentation. 145 00:15:27,880 --> 00:15:35,880 But they found this breakdown where you'd see most company... 146 00:15:35,880 --> 00:15:42,320 The majority of companies have fewer than 30 employees, so you would really consider 147 00:15:42,320 --> 00:15:45,120 that a small business. 148 00:15:45,120 --> 00:15:57,520 And then there are a handful of large companies with greater than 50 employees. 149 00:15:57,520 --> 00:16:08,560 But this is probably not the typical breakdown that you would see in a typical industry. 150 00:16:08,560 --> 00:16:11,480 Was somebody going to say something? 151 00:16:11,480 --> 00:16:20,880 But I'm just conjecturing, but I would imagine that in other industries, you may see employees 152 00:16:20,880 --> 00:16:24,000 typically working for large companies. 153 00:16:24,000 --> 00:16:26,120 But that's just a conjecture. 154 00:16:26,120 --> 00:16:31,520 I'd actually have to look at the data. 155 00:16:31,520 --> 00:16:39,240 So just to kind of keep moving on here with some of these other stats that I was breaking 156 00:16:39,240 --> 00:16:40,240 down. 157 00:16:40,240 --> 00:16:48,480 And before I get too far along here, I was just going to basically show you that if you 158 00:16:48,480 --> 00:16:56,400 go here to the GitHub, I've got all these stats. 159 00:16:56,400 --> 00:17:07,040 So here I'll put this link in the chat. 160 00:17:07,040 --> 00:17:14,740 And just to give you an idea of what this data looks like. 161 00:17:14,740 --> 00:17:25,820 This is old data, and so one of the subject matters of the day is refactoring, which is 162 00:17:25,820 --> 00:17:35,120 taking old code or not the best written code and rewriting it, cleaning it up, making it 163 00:17:35,120 --> 00:17:36,120 better. 164 00:17:36,120 --> 00:17:41,080 And in fact, that's often how you wind up with really, really good code, is by just 165 00:17:41,080 --> 00:17:43,840 cleaning up old code. 166 00:17:43,840 --> 00:17:51,440 So here we've got this mess of work that I did back in 2017. 167 00:17:51,440 --> 00:18:00,600 And so the idea is maybe it's time to refactor this, take this code, clean it up, get the 168 00:18:00,600 --> 00:18:08,200 data up to 2021, and it could be quite interesting. 169 00:18:08,200 --> 00:18:13,520 Long story short, we've got some cool data points here. 170 00:18:13,520 --> 00:18:23,360 So remember I was telling you, well, not that I was telling you, you can find these occupational 171 00:18:23,360 --> 00:18:35,000 licenses, this license data here. 172 00:18:35,000 --> 00:18:37,840 But you just have people's cities. 173 00:18:37,840 --> 00:18:51,240 So you have to actually do a bit of legwork to count the total number of employees in 174 00:18:51,240 --> 00:18:52,720 different places. 175 00:18:52,720 --> 00:18:59,520 And in fact, you may even have to go to the archives to get the actual month by month. 176 00:18:59,520 --> 00:19:05,280 So I'll start sharing some of these resources as I populate this to the recent date. 177 00:19:05,280 --> 00:19:13,600 But I'll have to share the links afterwards. 178 00:19:13,600 --> 00:19:21,520 But I've gone ahead here and month by month just created totals. 179 00:19:21,520 --> 00:19:29,120 So I've got total occupational licenses. 180 00:19:29,120 --> 00:19:31,880 There's a bunch of real cool data points. 181 00:19:31,880 --> 00:19:36,760 I probably wouldn't have formatted the keys like this today. 182 00:19:36,760 --> 00:19:40,240 Like snake case would probably be preferred. 183 00:19:40,240 --> 00:19:44,480 But this is what the data is. 184 00:19:44,480 --> 00:19:50,520 And so you've got all the licensees here. 185 00:19:50,520 --> 00:19:53,760 You have all the sales. 186 00:19:53,760 --> 00:20:01,800 And then you have some of these fields that I started to calculate, which are things like 187 00:20:01,800 --> 00:20:12,760 this is sort of occupational licenses per business license. 188 00:20:12,760 --> 00:20:23,760 So we don't know how many employees actually work at each particular business, which would 189 00:20:23,760 --> 00:20:30,600 be awesome because then we could get some really granular stats. 190 00:20:30,600 --> 00:20:37,720 But we can still just see the average number of licensees per license. 191 00:20:37,720 --> 00:20:46,720 And so what you see if you look at that number is you see it just in the early two years 192 00:20:46,720 --> 00:20:48,120 start to rise. 193 00:20:48,120 --> 00:20:53,680 So that's an indication that you're moving from small to large businesses. 194 00:20:53,680 --> 00:21:02,720 But still, you know, eleven employees per license is quite small. 195 00:21:02,720 --> 00:21:12,000 But OK, so we've got some good data here. 196 00:21:12,000 --> 00:21:19,480 You know, so now remember rule number one about data. 197 00:21:19,480 --> 00:21:21,560 Look at the data. 198 00:21:21,560 --> 00:21:28,920 So here is the growth of occupational licenses. 199 00:21:28,920 --> 00:21:34,200 So this is just how fast our license is growing. 200 00:21:34,200 --> 00:21:45,200 And so as you would expect when Colorado first started in 2014, it looks like. 201 00:21:45,200 --> 00:21:52,040 Of course, there's a lot of people entering high monthly growth rate of licenses, and 202 00:21:52,040 --> 00:21:56,440 then that gradually then that gradually diminishes. 203 00:21:56,440 --> 00:22:05,480 And so it would be really interesting to plot this out. 204 00:22:05,480 --> 00:22:13,040 You know, it would be real interesting to plot this out and see what's happening all 205 00:22:13,040 --> 00:22:18,240 the way up to, you know, twenty twenty one. 206 00:22:18,240 --> 00:22:22,640 So to that is sort of. 207 00:22:22,640 --> 00:22:26,880 The objective at hand. 208 00:22:26,880 --> 00:22:33,400 OK, and so. 209 00:22:33,400 --> 00:22:36,320 I was just going to show you a little bit of economics. 210 00:22:36,320 --> 00:22:40,840 I was sort of meaning to boil it down a little better. 211 00:22:40,840 --> 00:22:43,440 But. 212 00:22:43,440 --> 00:22:47,400 I'll touch on it, so don't let this scare you off or anything. 213 00:22:47,400 --> 00:22:53,080 But long story short is I was sort of going to do a little analysis here. 214 00:22:53,080 --> 00:22:57,280 And so. 215 00:22:57,280 --> 00:23:04,560 The way economic analysis really starts is you basically assume that. 216 00:23:04,560 --> 00:23:09,160 Things are produced given some sort of function. 217 00:23:09,160 --> 00:23:14,240 So right now, I'm basically saying, OK, so what's being produced? 218 00:23:14,240 --> 00:23:17,800 It's basically just cannabis flower. 219 00:23:17,800 --> 00:23:19,400 Right. 220 00:23:19,400 --> 00:23:26,040 So that's, you know, cannabis flower. 221 00:23:26,040 --> 00:23:28,840 And you know, it's dependent on something. 222 00:23:28,840 --> 00:23:29,840 Right. 223 00:23:29,840 --> 00:23:35,280 So I would just say, OK, well, that depends on the number of plants. 224 00:23:35,280 --> 00:23:44,920 Right. 225 00:23:44,920 --> 00:23:52,040 So so basically just saying, you know, you know, production, you know, is some sort of 226 00:23:52,040 --> 00:23:56,800 function of the number of plants. 227 00:23:56,800 --> 00:24:05,600 And what's nice is if you take the log of both sides, because you can apply any function 228 00:24:05,600 --> 00:24:15,720 to both sides, then you just get a nice linear function which you can estimate with the regression. 229 00:24:15,720 --> 00:24:28,840 So now you can just do an ordinary least squares regression of the log of flowers or sales 230 00:24:28,840 --> 00:24:33,480 on the log of plants. 231 00:24:33,480 --> 00:24:40,320 And you can, you know, you basically can estimate how. 232 00:24:40,320 --> 00:24:44,920 You know how productive the market is. 233 00:24:44,920 --> 00:24:54,680 And so this alpha is really is basically a measure of how productive the market is. 234 00:24:54,680 --> 00:25:03,000 If it's really close to zero, you're not going to have a very. 235 00:25:03,000 --> 00:25:05,120 It's going to be a really inefficient market. 236 00:25:05,120 --> 00:25:11,700 You're going to have to grow a ton of plants to just produce a tiny little bit of flower. 237 00:25:11,700 --> 00:25:19,000 But if alpha is close to one, then you don't, you know, you're it's a very efficient market 238 00:25:19,000 --> 00:25:26,400 and you're getting a lot of flower for the for the number of plants that you're growing. 239 00:25:26,400 --> 00:25:36,120 So that is the simplest, you know, production function. 240 00:25:36,120 --> 00:25:43,160 And so now we're going to introduce labor, right, because. 241 00:25:43,160 --> 00:25:46,000 You know, plants don't grow themselves. 242 00:25:46,000 --> 00:25:50,240 So here you just have plants and you have flower. 243 00:25:50,240 --> 00:25:57,080 OK, but now we're going to say, OK, we're now we're going to say, you know, we've got 244 00:25:57,080 --> 00:25:59,880 the same thing here. 245 00:25:59,880 --> 00:26:04,200 We've got our production. 246 00:26:04,200 --> 00:26:07,600 It depends on plants. 247 00:26:07,600 --> 00:26:12,440 But it also depends on labor. 248 00:26:12,440 --> 00:26:17,600 So you need plants or capital. 249 00:26:17,600 --> 00:26:22,400 So you know, K is essentially all physical goods. 250 00:26:22,400 --> 00:26:27,600 So just any anything physical required to produce. 251 00:26:27,600 --> 00:26:32,440 And then L is any any labor. 252 00:26:32,440 --> 00:26:34,800 So that is your managers. 253 00:26:34,800 --> 00:26:43,120 That is your you know, just your you know, your standard cultivators, your operators, 254 00:26:43,120 --> 00:26:49,320 all their time and energy, thought power, all that. 255 00:26:49,320 --> 00:26:55,200 And so, you know, we're basically saying, OK, you know, given that and, you know, the 256 00:26:55,200 --> 00:26:57,360 state and then. 257 00:26:57,360 --> 00:27:02,080 You know, you've got just things that are taken is given. 258 00:27:02,080 --> 00:27:03,080 So. 259 00:27:03,080 --> 00:27:10,280 So this is just, you know, your tech. 260 00:27:10,280 --> 00:27:14,360 You know, so that's just your tech. 261 00:27:14,360 --> 00:27:26,560 You know, and that's essentially your efficiency. 262 00:27:26,560 --> 00:27:29,680 So that's basically your efficiency of capital. 263 00:27:29,680 --> 00:27:38,240 And then, you know, beta is essentially your efficiency of. 264 00:27:38,240 --> 00:27:41,560 Of labor. 265 00:27:41,560 --> 00:27:54,520 So that's just sort of the mental framework, you know. 266 00:27:54,520 --> 00:27:58,720 So so that's just the mental framework that sort of show. 267 00:27:58,720 --> 00:28:00,800 OK, so. 268 00:28:00,800 --> 00:28:08,760 That's the framework for essentially why we're going to be plotting the data like this. 269 00:28:08,760 --> 00:28:14,680 And so so this is so basically, you know, this is three dimensions, right? 270 00:28:14,680 --> 00:28:16,600 So this is. 271 00:28:16,600 --> 00:28:18,600 Why your production? 272 00:28:18,600 --> 00:28:22,640 Hey, you know, your capital and L, your labor. 273 00:28:22,640 --> 00:28:24,720 So this is really a three dimension. 274 00:28:24,720 --> 00:28:28,800 You can think about that as like a three dimensional dome. 275 00:28:28,800 --> 00:28:35,280 And so this is just two two dimensions of that dome. 276 00:28:35,280 --> 00:28:40,560 And so here I've plotted. 277 00:28:40,560 --> 00:28:45,120 Occupational licenses, which would be our. 278 00:28:45,120 --> 00:28:46,720 Labor. 279 00:28:46,720 --> 00:28:52,880 Right, so this is basically, you know, LT down here. 280 00:28:52,880 --> 00:29:05,180 So right down here, we've got L. And then up here, we've got our production. 281 00:29:05,180 --> 00:29:08,880 So you know, as you would kind of expect. 282 00:29:08,880 --> 00:29:11,280 The more labor. 283 00:29:11,280 --> 00:29:12,280 That's in the market. 284 00:29:12,280 --> 00:29:17,680 You know, the more production you're going to have. 285 00:29:17,680 --> 00:29:28,240 And it's just sort of. You know, we're just beginning to to sort of form formalize a production 286 00:29:28,240 --> 00:29:29,240 function. 287 00:29:29,240 --> 00:29:31,680 OK, and so why? 288 00:29:31,680 --> 00:29:41,160 Well, the reason why is we're sort of trying to measure this curve here, right? 289 00:29:41,160 --> 00:29:42,600 Because. 290 00:29:42,600 --> 00:29:45,920 You know, we want to know. 291 00:29:45,920 --> 00:29:51,680 Because we want to know, OK, does that curve go like that? 292 00:29:51,680 --> 00:29:56,000 Does this curve go like that? 293 00:29:56,000 --> 00:29:57,280 Because. 294 00:29:57,280 --> 00:30:03,840 You know, right here, you know, you would have like, you know, you know, highly efficient 295 00:30:03,840 --> 00:30:13,240 labor and, you know, down here, you'd have, you know, real low efficient labor and. 296 00:30:13,240 --> 00:30:21,640 You know, essentially what economic theory tells us is, you know, the higher your efficiency 297 00:30:21,640 --> 00:30:30,620 of labor, you know, the higher wage you can expect to get. 298 00:30:30,620 --> 00:30:36,000 And so here's a similar chart with. 299 00:30:36,000 --> 00:30:37,800 The medical industry. 300 00:30:37,800 --> 00:30:45,400 And so as you as you see here, see here, you've got a real steep. 301 00:30:45,400 --> 00:30:50,320 Relation between licensees and retail revenue. 302 00:30:50,320 --> 00:30:52,320 And here. 303 00:30:52,320 --> 00:30:55,040 There's not quite a sharp relation. 304 00:30:55,040 --> 00:31:03,840 You've got fairly, fairly flat relation between the number of medical. 305 00:31:03,840 --> 00:31:07,720 Licenses and then medical revenue. 306 00:31:07,720 --> 00:31:14,840 And so all that would really mean is you would expect that the wage rate would actually be 307 00:31:14,840 --> 00:31:22,280 lower in the medical industry than it would be in the recreational industry. 308 00:31:22,280 --> 00:31:27,620 So that would that would just be a hypothesis that you would you could make. 309 00:31:27,620 --> 00:31:32,000 From looking at this data and, you know, using economic theory. 310 00:31:32,000 --> 00:31:34,920 But, you know, it's nothing more than a hypothesis. 311 00:31:34,920 --> 00:31:41,240 You know, you'd actually have to look at at the wage rates in the in the two in the two 312 00:31:41,240 --> 00:31:45,040 industries. 313 00:31:45,040 --> 00:31:51,240 So and then here's here's just the actual economic theory behind it. 314 00:31:51,240 --> 00:31:58,640 So basically. 315 00:31:58,640 --> 00:32:02,640 You know, economic theory would suggest that. 316 00:32:02,640 --> 00:32:06,800 You know, the competitive wage would be the one where you're basically getting paid, you 317 00:32:06,800 --> 00:32:09,080 know, your marginal product. 318 00:32:09,080 --> 00:32:16,120 So what that says in simplest terms is. 319 00:32:16,120 --> 00:32:21,440 You're going to produce a certain amount of value for your for your work. 320 00:32:21,440 --> 00:32:29,160 You know, no one would pay you more than the value you you generate. 321 00:32:29,160 --> 00:32:37,320 But if they're paying you less than the value you generate in a perfectly competitive market, 322 00:32:37,320 --> 00:32:43,840 somebody else would come along and say, hey, somebody's paying him less than the value 323 00:32:43,840 --> 00:32:51,040 he's producing or she, then I'm going to, you know, pay them just a little bit more. 324 00:32:51,040 --> 00:32:57,720 And so that would just keep happening until you're essentially paid the exact amount 325 00:32:57,720 --> 00:33:00,440 of value that you're producing. 326 00:33:00,440 --> 00:33:02,520 So it's all sort of conceptual. 327 00:33:02,520 --> 00:33:09,840 And, you know, you would expect that, you know, in the real world, you wouldn't expect 328 00:33:09,840 --> 00:33:15,600 that anybody would be getting paid their competitive wage just because of market frictions and 329 00:33:15,600 --> 00:33:17,860 all sorts of factors. 330 00:33:17,860 --> 00:33:26,720 So this is an entirely theoretical wage, but it's just a benchmark that we can just use 331 00:33:26,720 --> 00:33:30,280 to, you know, start estimating. 332 00:33:30,280 --> 00:33:31,280 OK. 333 00:33:31,280 --> 00:33:36,960 You know, OK, we don't know people's wages. 334 00:33:36,960 --> 00:33:41,120 All we know is retail revenue and the licenses. 335 00:33:41,120 --> 00:33:50,080 You know, is there any way we can estimate their wages, you know, given that data? 336 00:33:50,080 --> 00:34:01,120 Well, if you go back to our production function here, so here we have ykT. 337 00:34:01,120 --> 00:34:17,480 Basically we're just going to add, you know, lT to the beta right there. 338 00:34:17,480 --> 00:34:23,880 And then that gets log linearized down here. 339 00:34:23,880 --> 00:34:32,700 And that turns into, you know, beta log lT. 340 00:34:32,700 --> 00:34:42,080 So we still have our linear equation where we're just going to take a regression of production, 341 00:34:42,080 --> 00:34:44,320 which I'll just call sales. 342 00:34:44,320 --> 00:34:46,040 So that'll just be revenue. 343 00:34:46,040 --> 00:34:52,720 So we're just going to be regressing revenue on some measure of capital, which I'm calling 344 00:34:52,720 --> 00:35:02,480 plants and then on labor, the log of labor, which I'm calling total occupational licenses. 345 00:35:02,480 --> 00:35:09,280 So you know, it's all just a crude approximation and just, you know, just seeing what we can 346 00:35:09,280 --> 00:35:10,280 make fit. 347 00:35:10,280 --> 00:35:22,160 It's just interesting to just sort of make this visual conceptual model of the industry. 348 00:35:22,160 --> 00:35:33,840 So we're saying, okay, the wage, that was, we can probably estimate it to be competitive. 349 00:35:33,840 --> 00:35:38,120 Also in the, so this is just more data. 350 00:35:38,120 --> 00:35:40,120 This probably should have belonged earlier. 351 00:35:40,120 --> 00:35:46,360 This is just a breakdown of actual positions held. 352 00:35:46,360 --> 00:35:48,520 Okay. 353 00:35:48,520 --> 00:36:00,640 So we've done the nitty gritty. 354 00:36:00,640 --> 00:36:03,600 Now we want to calculate this. 355 00:36:03,600 --> 00:36:04,960 Okay. 356 00:36:04,960 --> 00:36:20,160 So I'm going to real quick show you how that would be done. 357 00:36:20,160 --> 00:36:34,080 So you're going to import this data here where you've got total occupational licenses. 358 00:36:34,080 --> 00:36:37,200 And we also have total revenue. 359 00:36:37,200 --> 00:36:43,400 So that's basically, this is basically going to be our Y and then this is basically going 360 00:36:43,400 --> 00:36:46,400 to be our L. 361 00:36:46,400 --> 00:37:04,320 And I believe we have the number of plants. 362 00:37:04,320 --> 00:37:05,320 I think yes. 363 00:37:05,320 --> 00:37:09,360 So here I should have average cultivated plants. 364 00:37:09,360 --> 00:37:14,800 So let's, you know, read in that data. 365 00:37:14,800 --> 00:37:19,200 Hopefully everything's there. 366 00:37:19,200 --> 00:37:21,920 Number one rule about data. 367 00:37:21,920 --> 00:37:23,280 Look at the data. 368 00:37:23,280 --> 00:37:27,760 So this is total revenue. 369 00:37:27,760 --> 00:37:33,600 Oops. 370 00:37:33,600 --> 00:37:51,760 Here we have total plants, which will basically be our KT. 371 00:37:51,760 --> 00:37:54,320 So we've got our YT, KT and then our LT. 372 00:37:54,320 --> 00:37:57,920 And so we've got our total labor. 373 00:37:57,920 --> 00:38:03,400 Also and this is what I was wanting to talk about, refactoring. 374 00:38:03,400 --> 00:38:10,080 This is old, bad code that needs to be refactored. 375 00:38:10,080 --> 00:38:15,680 So I wouldn't really actually recommend labeling your variables like this. 376 00:38:15,680 --> 00:38:22,160 So like for Python, you know, you should probably go, you know, total labor. 377 00:38:22,160 --> 00:38:24,960 So this code needs to be refactored. 378 00:38:24,960 --> 00:38:28,080 So that is the name. 379 00:38:28,080 --> 00:38:30,360 That is the buzzword of the day. 380 00:38:30,360 --> 00:38:37,280 And so this code needs to be refactored. 381 00:38:37,280 --> 00:38:50,440 Remember earlier I said it's real interesting to supplement your data with external data 382 00:38:50,440 --> 00:38:51,440 sources. 383 00:38:51,440 --> 00:39:03,760 Well, here I'm actually using the Federal Reserve's thread API. 384 00:39:03,760 --> 00:39:10,960 And so I'm not certain I have this. 385 00:39:10,960 --> 00:39:13,960 Okay. 386 00:39:13,960 --> 00:39:25,680 So we'll need to install the thread Python API. 387 00:39:25,680 --> 00:39:42,120 And so let's actually just, you know, see if we can't do that real quick. 388 00:39:42,120 --> 00:40:00,360 So I think we can just do... 389 00:40:00,360 --> 00:40:13,960 And so what I'm using here is basically the Federal Reserve has, you know, this would 390 00:40:13,960 --> 00:40:21,000 be the go to place if you just need monthly, yearly, you know, potentially even more frequently, 391 00:40:21,000 --> 00:40:26,080 but typically the not so frequent series. 392 00:40:26,080 --> 00:40:33,920 So you know, anything from, you know, so this is where a lot of economists get their data. 393 00:40:33,920 --> 00:40:42,320 Honestly I probably under utilize it because there is a bunch of awesome data here, but 394 00:40:42,320 --> 00:40:49,880 essentially right now pulling the... 395 00:40:49,880 --> 00:40:53,080 Right? 396 00:40:53,080 --> 00:40:55,680 Because we're estimating the wage rate. 397 00:40:55,680 --> 00:41:02,840 So it wouldn't be any fun to just pull the average wage data. 398 00:41:02,840 --> 00:41:10,200 But basically what I'm pulling here is just the... 399 00:41:10,200 --> 00:41:16,280 This is just the average number of hours a week that somebody would work. 400 00:41:16,280 --> 00:41:25,240 Because remember earlier I said, oh, I'm just in a conjecture that they work 40 hours per 401 00:41:25,240 --> 00:41:26,240 week. 402 00:41:26,240 --> 00:41:30,440 But then I started thinking, you know, that's not so realistic. 403 00:41:30,440 --> 00:41:37,280 And so what you can do is you can actually, you know, you can actually get the average 404 00:41:37,280 --> 00:41:39,880 hourly work week. 405 00:41:39,880 --> 00:41:45,720 So you know, so like you just slowly, you know, getting your approximations closer and 406 00:41:45,720 --> 00:41:49,000 closer to realistic. 407 00:41:49,000 --> 00:41:55,400 So as you see, it's not, people aren't working 40 hours a week on average. 408 00:41:55,400 --> 00:42:03,480 It's typically, you know, it's close to, you know, 30, 33, 34 or so. 409 00:42:03,480 --> 00:42:04,480 Okay. 410 00:42:04,480 --> 00:42:17,080 And so this is old code and let's see if we can still get this to work. 411 00:42:17,080 --> 00:42:30,920 So it looks like we need numpy. 412 00:42:30,920 --> 00:42:38,440 And we need stats models. 413 00:42:38,440 --> 00:42:58,360 And so essentially what I'm doing here is just getting everything ready for this regression. 414 00:42:58,360 --> 00:43:07,160 I'm just getting the log of y, the log of capital, and the log of labor, as well as 415 00:43:07,160 --> 00:43:14,920 a constant, just to represent technology. 416 00:43:14,920 --> 00:43:19,240 And this is where we're going to start off-roading a little bit. 417 00:43:19,240 --> 00:43:24,960 I need to probably touch up this code. 418 00:43:24,960 --> 00:43:32,560 So I'm not even going to promise that it's going to work here, but we can at least give 419 00:43:32,560 --> 00:43:39,200 it a shot. 420 00:43:39,200 --> 00:43:43,280 See it was even written in Python too. 421 00:43:43,280 --> 00:43:51,800 So we're going to do a regression here. 422 00:43:51,800 --> 00:43:54,520 Okay. 423 00:43:54,520 --> 00:44:01,920 And so remember, we barely even have any data. 424 00:44:01,920 --> 00:44:03,520 Still we have some data. 425 00:44:03,520 --> 00:44:06,440 We can get a lot more data. 426 00:44:06,440 --> 00:44:12,500 So we only have, you know, 29 observations. 427 00:44:12,500 --> 00:44:15,760 So it would be much better to run this through 2021. 428 00:44:15,760 --> 00:44:21,260 And look, we're estimating these exact things. 429 00:44:21,260 --> 00:44:25,680 So we estimated our constant. 430 00:44:25,680 --> 00:44:34,320 So that's, so right, so we just estimated that term. 431 00:44:34,320 --> 00:44:51,880 So we just, we just estimated that, this whole term. 432 00:44:51,880 --> 00:45:07,480 And we estimated x1, which is alpha. 433 00:45:07,480 --> 00:45:20,800 So we write, oops. 434 00:45:20,800 --> 00:45:28,920 So we just estimated alpha. 435 00:45:28,920 --> 00:45:34,960 And we estimated that to be 0.28. 436 00:45:34,960 --> 00:45:51,120 And then we estimated beta to be 0.0684. 437 00:45:51,120 --> 00:46:01,800 And if you look at this, if you add these two coefficients together, well, it's not quite 438 00:46:01,800 --> 00:46:14,160 one, but essentially economic theory would suggest that alpha plus beta would equal one. 439 00:46:14,160 --> 00:46:21,360 So we're obviously leaving some explanation left on the table, right? 440 00:46:21,360 --> 00:46:27,200 There's still about 10% of efficiency that's not really being explained here. 441 00:46:27,200 --> 00:46:37,440 But, you know, we're still explaining, you know, a lot of, you know, alpha and beta, 442 00:46:37,440 --> 00:46:44,840 you know, but they're close to what they should be, you know, theoretically. 443 00:46:44,840 --> 00:46:53,560 And so, so remember, so this is just a measure of efficiency. 444 00:46:53,560 --> 00:46:58,640 So remember, the closer to one it is, the more efficient it is. 445 00:46:58,640 --> 00:47:09,200 So here we've got fairly efficient labor and then capital is not as efficient. 446 00:47:09,200 --> 00:47:17,560 And remember, if we go back here, so we now have beta. 447 00:47:17,560 --> 00:47:29,240 Remember, so beta is, you know, remember beta is like about 0.6 or so. 448 00:47:29,240 --> 00:47:35,120 So now we can actually calculate the competitive wage. 449 00:47:35,120 --> 00:47:43,560 So now at any given point in time, we can say, okay, what is the competitive wage? 450 00:47:43,560 --> 00:47:44,560 Right? 451 00:47:44,560 --> 00:47:57,320 So it's just going to be the total amount of sales divided by the total amount of labor, 452 00:47:57,320 --> 00:48:03,480 you know, times 0.6. 453 00:48:03,480 --> 00:48:26,520 So we now have the competitive wage or an estimate. 454 00:48:26,520 --> 00:48:30,400 So this is, let me really benchmark that there. 455 00:48:30,400 --> 00:48:37,560 This is, you know, essentially, you know, a complete estimate, you know, I mean, keep 456 00:48:37,560 --> 00:48:42,280 in mind all of the assumptions we've made along the way. 457 00:48:42,280 --> 00:48:53,160 So we first assumed that production is a function of plants and labor. 458 00:48:53,160 --> 00:49:03,200 Obviously, production is more than a function of plants and simply measuring labor's total 459 00:49:03,200 --> 00:49:09,160 occupational licenses is also sort of a big mental stretch. 460 00:49:09,160 --> 00:49:13,720 So we did some mental gymnastics there. 461 00:49:13,720 --> 00:49:20,760 And just in fact, the assumption that this is what the production function looks like, 462 00:49:20,760 --> 00:49:25,760 that's also an enormous assumption. 463 00:49:25,760 --> 00:49:35,320 And then we assumed that everything's perfectly competitive. 464 00:49:35,320 --> 00:49:43,960 And you know, obviously, you know, there's a lot of factors, you know, that come into 465 00:49:43,960 --> 00:49:45,960 play there. 466 00:49:45,960 --> 00:49:53,880 And so the long story short, lots of considerations, but we're estimating, you know, a competitive 467 00:49:53,880 --> 00:50:05,000 wage, you know, is fluctuating somewhere between, you know, $9 and $11 per hour. 468 00:50:05,000 --> 00:50:10,600 And so, you know, like I said, this is an estimate. 469 00:50:10,600 --> 00:50:16,040 And we actually can put confidence intervals on that. 470 00:50:16,040 --> 00:50:31,480 And I think I've done that. 471 00:50:31,480 --> 00:50:34,520 Let's try to plot this all in one. 472 00:50:34,520 --> 00:50:44,640 So hold on, I'll be bringing this home here. 473 00:50:44,640 --> 00:50:51,920 All right. 474 00:50:51,920 --> 00:50:57,960 And so here, you know, you could plot this in a more fanciful way. 475 00:50:57,960 --> 00:51:04,160 But here, we're essentially saying, OK, you know, we think, you know, the real wage, you 476 00:51:04,160 --> 00:51:13,520 know, could be anywhere between, you know, $4 and, you know, $18 or so per hour. 477 00:51:13,520 --> 00:51:21,520 And so, you know, with our best guess at around, you know, $10 an hour. 478 00:51:21,520 --> 00:51:26,900 And so, you know, this is sort of bringing, you know, bringing it home. 479 00:51:26,900 --> 00:51:39,680 So basically, given, you know, all we were given was this data set here, just the Colorado 480 00:51:39,680 --> 00:51:47,080 data publicly available, you know, as well as, you know, the Fred API. 481 00:51:47,080 --> 00:51:55,560 We were able to create monthly totals. 482 00:51:55,560 --> 00:52:10,360 Then we were able to use economic theory to estimate what a competitive wage may be. 483 00:52:10,360 --> 00:52:15,680 And then we estimated it and then we plotted it. 484 00:52:15,680 --> 00:52:22,600 And now, you know, now people can know that, hey, this is what, you know, this is what 485 00:52:22,600 --> 00:52:26,600 the competitive wage is in the cannabis industry. 486 00:52:26,600 --> 00:52:32,400 And I mean, if you're looking at like a competitive wage of, you know, up to, you know, $18 or 487 00:52:32,400 --> 00:52:45,520 so an hour, well, that may be why there are a lot of legal cannabis jobs, because you 488 00:52:45,520 --> 00:52:51,160 have a high, a high competitive wage. 489 00:52:51,160 --> 00:53:01,520 So, I think that brings us to the end of the presentation. 490 00:53:01,520 --> 00:53:10,120 I kind of wanted to save a little bit more time for questions, but, you know, that's 491 00:53:10,120 --> 00:53:12,480 some of the work I'm doing. 492 00:53:12,480 --> 00:53:21,760 And now the task is to rinse and repeat with all the data up to 2021. 493 00:53:21,760 --> 00:53:25,440 So, thanks. 494 00:53:25,440 --> 00:53:26,440 That was really cool. 495 00:53:26,440 --> 00:53:28,680 But I have another meeting I have to go to. 496 00:53:28,680 --> 00:53:29,680 Okay. 497 00:53:29,680 --> 00:53:30,680 So, I'll see you next week. 498 00:53:30,680 --> 00:53:31,680 Definitely. 499 00:53:31,680 --> 00:53:37,600 But I'll go ahead and conclude here and I'll get this recording up, get some of the material 500 00:53:37,600 --> 00:53:38,920 up. 501 00:53:38,920 --> 00:53:45,360 And I know it was a lot to take in, but it was even a lot for me to take in. 502 00:53:45,360 --> 00:53:48,200 I've sort of got to get refreshed on it. 503 00:53:48,200 --> 00:53:52,160 But, you know, like I said, that's what I'm working on. 504 00:53:52,160 --> 00:53:55,160 It's got a lot of cleaning up to do. 505 00:53:55,160 --> 00:53:59,120 But there's some interesting insights there to be had. 506 00:53:59,120 --> 00:54:00,120 Yeah, definitely. 507 00:54:00,120 --> 00:54:01,120 Thank you. 508 00:54:01,120 --> 00:54:03,120 Thank you for presenting that. 509 00:54:03,120 --> 00:54:05,320 You're very welcome, Alice. 510 00:54:05,320 --> 00:54:06,640 Yeah, there you go. 511 00:54:06,640 --> 00:54:07,880 So, thank you for coming. 512 00:54:07,880 --> 00:54:09,840 I hope you got something out of it. 513 00:54:09,840 --> 00:54:12,640 Do you have a little bit of time for a question? 514 00:54:12,640 --> 00:54:14,840 Definitely, by all means. 515 00:54:14,840 --> 00:54:17,960 If you were in, I don't know anything about economy. 516 00:54:17,960 --> 00:54:20,160 So that was a new piece for me. 517 00:54:20,160 --> 00:54:26,160 But if you were in a market where things are stabilized and you don't have that many more, 518 00:54:26,160 --> 00:54:33,440 like you have more license or less license, like it's very thin variation. 519 00:54:33,440 --> 00:54:36,440 And let's say the revenue is more or less the same. 520 00:54:36,440 --> 00:54:46,080 And in that plot, the reason you were able to plot revenue versus occupational license 521 00:54:46,080 --> 00:54:51,720 and to have this long line is because the market had been growing rapidly, right? 522 00:54:51,720 --> 00:54:52,720 Exactly. 523 00:54:52,720 --> 00:55:04,200 And so if you were in a stable market, you'd have a much smaller plot, per se. 524 00:55:04,200 --> 00:55:08,280 So, exactly, and you hit on something interesting there. 525 00:55:08,280 --> 00:55:17,280 So with statistics, variation is key if you're trying to get some good measurements. 526 00:55:17,280 --> 00:55:22,560 So it will be interesting to see. 527 00:55:22,560 --> 00:55:30,000 And that's why it's so worthwhile to extend this out to the current day is, okay, did 528 00:55:30,000 --> 00:55:34,480 the occupational licenses, did that level off? 529 00:55:34,480 --> 00:55:39,200 And if so, what's the effect on the market? 530 00:55:39,200 --> 00:55:41,480 Did the market keep growing? 531 00:55:41,480 --> 00:55:48,920 Because if the market keeps growing and then the occupational licenses level off, you'd 532 00:55:48,920 --> 00:55:53,080 expect those people maybe to get paid, maybe get paid more. 533 00:55:53,080 --> 00:55:57,400 Maybe they're becoming better at their jobs, becoming more efficient or something like 534 00:55:57,400 --> 00:55:58,400 that. 535 00:55:58,400 --> 00:55:59,400 Yeah. 536 00:55:59,400 --> 00:56:00,400 That's interesting. 537 00:56:00,400 --> 00:56:03,720 Yeah, that's what I'm thinking. 538 00:56:03,720 --> 00:56:06,000 And so, yeah, so it's exciting. 539 00:56:06,000 --> 00:56:13,480 And so I'd wanted to extend this out to 2021 for today, but there's a good bit of data 540 00:56:13,480 --> 00:56:15,360 wrangling ahead. 541 00:56:15,360 --> 00:56:21,520 So that's why I wanted to go ahead and share what I had with you today. 542 00:56:21,520 --> 00:56:26,560 And the data that you had for 2014 and 2015, you had in a spreadsheet. 543 00:56:26,560 --> 00:56:29,680 What was something that you got from a different data source? 544 00:56:29,680 --> 00:56:33,240 And then the links you shared with that. 545 00:56:33,240 --> 00:56:36,760 So you see the GitHub in the chat there? 546 00:56:36,760 --> 00:56:37,760 Yeah. 547 00:56:37,760 --> 00:56:44,620 So I'm so there is an Excel spreadsheet there. 548 00:56:44,620 --> 00:56:49,840 You may have to essentially clone the GitHub if you want. 549 00:56:49,840 --> 00:56:56,760 I could send you my email afterwards and then essentially just email you that data set if 550 00:56:56,760 --> 00:56:58,760 you would like. 551 00:56:58,760 --> 00:57:00,120 Yeah. 552 00:57:00,120 --> 00:57:05,040 For me right now, it's more understanding kind of like where you so I see the Excel. 553 00:57:05,040 --> 00:57:13,360 I mean, I see this CSV file on today's presentation data directory. 554 00:57:13,360 --> 00:57:23,320 But you're saying that to get data up until 2021, then you'd need to do like get the data 555 00:57:23,320 --> 00:57:25,320 actually from those websites. 556 00:57:25,320 --> 00:57:26,320 Exactly. 557 00:57:26,320 --> 00:57:30,840 And so here, let me put that in the chat here. 558 00:57:30,840 --> 00:57:40,200 But essentially, when I did it, I was looking at archived Excel versions. 559 00:57:40,200 --> 00:57:44,520 You may be able to do it with that API I was sharing earlier. 560 00:57:44,520 --> 00:57:50,280 But now that I'm starting to think about it, I think that may just be at least for the 561 00:57:50,280 --> 00:57:54,160 occupational licenses, that may only be a current snapshot. 562 00:57:54,160 --> 00:58:00,160 So I may have to share with you the links to the actual archives. 563 00:58:00,160 --> 00:58:06,040 So I actually do that is actually sort of a gap. 564 00:58:06,040 --> 00:58:13,000 So I realized that in my preparation today that I was missing that data source. 565 00:58:13,000 --> 00:58:15,560 So I still have to scrounge that up. 566 00:58:15,560 --> 00:58:19,840 But I'll send that to you when I get it. 567 00:58:19,840 --> 00:58:23,640 Because I've got it somewhere. 568 00:58:23,640 --> 00:58:24,640 I've got the data. 569 00:58:24,640 --> 00:58:32,560 I just need to find out where I got that online again, so we can get it through 2021. 570 00:58:32,560 --> 00:58:41,480 And on the GitHub, you had a bunch of handouts. 571 00:58:41,480 --> 00:58:46,080 What were you wondering? 572 00:58:46,080 --> 00:58:48,440 Sorry, there's a little... 573 00:58:48,440 --> 00:58:50,920 I hear myself on your end. 574 00:58:50,920 --> 00:58:51,920 Sorry. 575 00:58:51,920 --> 00:58:52,920 Ah, here. 576 00:58:52,920 --> 00:58:59,640 I can mute myself when I'm speaking. 577 00:58:59,640 --> 00:59:08,520 I was saying in the... so I'm on the March 10 repository of your GitHub. 578 00:59:08,520 --> 00:59:11,080 And there were a bunch of handouts. 579 00:59:11,080 --> 00:59:16,040 And I was curious about mapping basics with bucket. 580 00:59:16,040 --> 00:59:20,280 Like you put the links to articles. 581 00:59:20,280 --> 00:59:22,720 What was the idea for you there? 582 00:59:22,720 --> 00:59:23,840 Oh, yes. 583 00:59:23,840 --> 00:59:31,760 So part of what I'm doing here is I'm just sort of keeping track of just interesting 584 00:59:31,760 --> 00:59:38,920 data science slash cannabis data that I come across on a week-to-week basis. 585 00:59:38,920 --> 00:59:41,520 I don't know if it's helpful or not. 586 00:59:41,520 --> 00:59:47,080 But I just kind of thought that I just was starting to just accumulate things for the 587 00:59:47,080 --> 00:59:48,080 group. 588 00:59:48,080 --> 00:59:53,800 And I just kind of say like, hey, these are things I found in the past week. 589 00:59:53,800 --> 00:59:55,920 I think they're interesting. 590 00:59:55,920 --> 00:59:56,920 They're worth looking at. 591 00:59:56,920 --> 01:00:05,280 I'm just going to share them with the group because I've got, you know, 1,001 pies in 592 01:00:05,280 --> 01:00:06,280 the oven. 593 01:00:06,280 --> 01:00:13,440 And I may not get to all these awesome ideas and projects. 594 01:00:13,440 --> 01:00:20,120 And so I found that instead of just leaving them to get dusty, I'll just share them because 595 01:00:20,120 --> 01:00:25,640 like, for example, Charles was finding some use for the data. 596 01:00:25,640 --> 01:00:30,960 And so instead of them just getting dust, I'm just going to share them. 597 01:00:30,960 --> 01:00:33,040 And I'll get to them eventually. 598 01:00:33,040 --> 01:00:38,640 And then if you want to poke around at them for the time being, have at it. 599 01:00:38,640 --> 01:00:42,080 So that's great. 600 01:00:42,080 --> 01:00:48,320 So you have past week's presentations in that GitHub. 601 01:00:48,320 --> 01:00:52,500 So the actual videos, I'm still uploading online. 602 01:00:52,500 --> 01:00:59,920 So I'm going to I'm basically creating a web page where I'll just have all the video archives. 603 01:00:59,920 --> 01:01:03,680 Like I said, it's on my to do list. 604 01:01:03,680 --> 01:01:07,000 And in fact, it's on my today's to do list. 605 01:01:07,000 --> 01:01:17,200 So you should check your messages on Meetup and I'll make sure to I'll send you the links 606 01:01:17,200 --> 01:01:21,400 because those will be coming out the pipeline real soon. 607 01:01:21,400 --> 01:01:22,400 Yeah. 608 01:01:22,400 --> 01:01:29,240 Do you have some time to talk just in general about data science? 609 01:01:29,240 --> 01:01:30,240 I don't know. 610 01:01:30,240 --> 01:01:31,240 I don't want to take too much. 611 01:01:31,240 --> 01:01:34,240 Oh, I definitely do. 612 01:01:34,240 --> 01:01:36,960 I'm always happy to talk about data science. 613 01:01:36,960 --> 01:01:41,960 Would you want to like schedule like a one on one? 614 01:01:41,960 --> 01:01:42,960 Maybe. 615 01:01:42,960 --> 01:01:47,520 Maybe it'd be better if you like you have it in your schedule and we can we can chat. 616 01:01:47,520 --> 01:01:50,320 I'm an aspiring data scientist. 617 01:01:50,320 --> 01:01:51,320 Awesome. 618 01:01:51,320 --> 01:01:55,880 I'm pivoting from physics research. 619 01:01:55,880 --> 01:02:03,600 And I've enrolled into a data science boot camp that I'm doing in a couple of months. 620 01:02:03,600 --> 01:02:07,680 And one of the things that they ask us is to build a project. 621 01:02:07,680 --> 01:02:13,360 And so I'm kind of looking at different things that like different data sets that people 622 01:02:13,360 --> 01:02:16,080 haven't necessarily looked at a lot. 623 01:02:16,080 --> 01:02:19,800 And kind of his data seems like really rich. 624 01:02:19,800 --> 01:02:23,740 And actually, I don't know that that many people have looked at it. 625 01:02:23,740 --> 01:02:28,400 And so I'm thinking that that could become my project. 626 01:02:28,400 --> 01:02:34,840 If that was the case, then I would work on it probably like 10, 20 hours per week on 627 01:02:34,840 --> 01:02:35,840 it. 628 01:02:35,840 --> 01:02:42,440 And so I'm trying to see whether it's a good opportunity for me to start getting into. 629 01:02:42,440 --> 01:02:43,440 Definitely Alice. 630 01:02:43,440 --> 01:02:48,760 Well, I can definitely point you in the direction of some good resources, because like you said, 631 01:02:48,760 --> 01:02:56,720 it may be unparalleled, the amount of public data, because very few industries are actually 632 01:02:56,720 --> 01:03:02,200 regulated on such a granular level. 633 01:03:02,200 --> 01:03:08,480 So they are recording data at every stage. 634 01:03:08,480 --> 01:03:16,720 And the public has kind of opted towards a public data approach towards cannabis, because 635 01:03:16,720 --> 01:03:22,640 you just kind of want to let people know about it, know that it's going on, measure it. 636 01:03:22,640 --> 01:03:29,320 And it's new, so people were able to set up, in some cases, APIs. 637 01:03:29,320 --> 01:03:33,040 So there are some awesome sources. 638 01:03:33,040 --> 01:03:40,280 And like you said, it's basically a firehose of data. 639 01:03:40,280 --> 01:03:43,000 And people are just scrambling at it. 640 01:03:43,000 --> 01:03:45,360 And so there's a lot to be uncovered. 641 01:03:45,360 --> 01:03:46,360 OK. 642 01:03:46,360 --> 01:03:47,360 Yeah, yeah. 643 01:03:47,360 --> 01:03:50,880 It's really, really exciting. 644 01:03:50,880 --> 01:03:59,760 I'm going to try to come through a little bit of your GitHub repo for the past presentations 645 01:03:59,760 --> 01:04:02,200 that you've done, kind of trying to understand. 646 01:04:02,200 --> 01:04:04,080 Right now, I can see there's a lot of data. 647 01:04:04,080 --> 01:04:09,560 I'm still a little bit unclear on what are the most interesting questions to focus on. 648 01:04:09,560 --> 01:04:11,920 And I guess that's your real house. 649 01:04:11,920 --> 01:04:12,920 Ooh. 650 01:04:12,920 --> 01:04:16,760 Well, the warning, the repository is a bit of a mess right now. 651 01:04:16,760 --> 01:04:17,760 It needs a bit of cleanup. 652 01:04:17,760 --> 01:04:21,480 I've been just posting things for the time being. 653 01:04:21,480 --> 01:04:24,880 But if you'd like, I could think of some things. 654 01:04:24,880 --> 01:04:33,640 And then when we speak, we could hammer out some good projects that you may want to poke 655 01:04:33,640 --> 01:04:34,640 around at. 656 01:04:34,640 --> 01:04:35,640 That sounds great. 657 01:04:35,640 --> 01:04:45,400 Do you know what your availabilities are in the next week or next week? 658 01:04:45,400 --> 01:04:48,280 I'm always available afternoon. 659 01:04:48,280 --> 01:04:54,120 And then on Fridays, I'm usually wide open. 660 01:04:54,120 --> 01:04:59,280 So how about tomorrow afternoon? 661 01:04:59,280 --> 01:05:00,280 Definitely. 662 01:05:00,280 --> 01:05:01,280 OK. 663 01:05:01,280 --> 01:05:04,520 Do you have a time that's best for you? 664 01:05:04,520 --> 01:05:05,520 2 PM. 665 01:05:05,520 --> 01:05:10,680 And tomorrow is Thursday. 666 01:05:10,680 --> 01:05:18,560 So I do have an engagement at 2.30 on tomorrow. 667 01:05:18,560 --> 01:05:25,520 So if we can do Friday. 668 01:05:25,520 --> 01:05:30,080 How about I'm going to send you an email and you think of the best time for you. 669 01:05:30,080 --> 01:05:33,640 I'm going to give you my time windows when I'm free and then you pick the time. 670 01:05:33,640 --> 01:05:34,640 Awesome. 671 01:05:34,640 --> 01:05:35,640 Awesome. 672 01:05:35,640 --> 01:05:36,640 All right. 673 01:05:36,640 --> 01:05:37,640 Well, this is exciting, Alice. 674 01:05:37,640 --> 01:05:38,640 It was awesome meeting you. 675 01:05:38,640 --> 01:05:41,720 Can you put your email in the chat? 676 01:05:41,720 --> 01:05:42,720 Yes. 677 01:05:42,720 --> 01:05:50,680 So you can just reach me at just Keegan Ski at Gmail. 678 01:05:50,680 --> 01:05:51,680 All right. 679 01:05:51,680 --> 01:05:54,440 Well, it was great. 680 01:05:54,440 --> 01:05:58,440 I wanted to attend this like, I don't know, three, four weeks ago and I couldn't. 681 01:05:58,440 --> 01:06:01,120 I'm really glad I got to do that. 682 01:06:01,120 --> 01:06:05,600 Well, today ended up being an interesting presentation. 683 01:06:05,600 --> 01:06:11,080 There was a lot in there, but I thought it was an interesting one. 684 01:06:11,080 --> 01:06:13,400 But I think it still needs to get cleaned up. 685 01:06:13,400 --> 01:06:18,600 Well, you get the content and then you had the data science and the coding. 686 01:06:18,600 --> 01:06:21,960 And so that was that was a lot to put in like one hour. 687 01:06:21,960 --> 01:06:26,000 But I definitely like the approach and like knowing a little bit more background about 688 01:06:26,000 --> 01:06:28,040 like the economical science behind it. 689 01:06:28,040 --> 01:06:30,080 So that was really interesting to me. 690 01:06:30,080 --> 01:06:32,120 Oh, well, all right. 691 01:06:32,120 --> 01:06:33,600 I'll send you an email right away. 692 01:06:33,600 --> 01:06:34,600 Nice. 693 01:06:34,600 --> 01:06:36,160 I'm glad you got something out of it, Alice. 694 01:06:36,160 --> 01:06:37,960 So I'll look forward to speaking with you. 695 01:06:37,960 --> 01:06:38,960 All right. 696 01:06:38,960 --> 01:06:39,960 Bye. 697 01:06:39,960 --> 01:06:40,960 Have an awesome day. 698 01:06:40,960 --> 01:07:04,560 Thank you.