WEBVTT NOTE This file was generated by Descript 00:00:00.030 --> 00:00:03.270 Judith: Welcome to Berry's In the Interim podcast, where we explore the 00:00:03.270 --> 00:00:07.830 cutting edge of innovative clinical trial design for the pharmaceutical and 00:00:07.830 --> 00:00:10.170 medical industries, and so much more. 00:00:11.040 --> 00:00:11.730 Let's dive in. 00:00:13.081 --> 00:00:13.561 Scott Berry: All right. 00:00:13.561 --> 00:00:14.791 Welcome everybody. 00:00:14.851 --> 00:00:20.041 Uh, back to in the interim, uh, I'm Scott Berry, uh, at Berry Consultants 00:00:20.041 --> 00:00:22.561 and I have a really cool topic for today. 00:00:22.891 --> 00:00:28.621 We're gonna talk about, uh, Leonard Jimmy Savage, and we have. 00:00:29.671 --> 00:00:33.121 Perhaps the best person to talk to it, who's also, by the way, my 00:00:33.121 --> 00:00:38.641 father, um, and weird relations, almost a brother and a father. 00:00:38.641 --> 00:00:43.771 But we'll get into that, uh, biological father and an academic brother, uh, 00:00:43.831 --> 00:00:46.861 uh, in this, so, so Don, welcome. 00:00:46.861 --> 00:00:49.561 And, uh, first question. 00:00:49.561 --> 00:00:50.071 Leonard. 00:00:50.101 --> 00:00:51.661 Leonard, Jimmy Savage. 00:00:51.991 --> 00:00:53.371 I feel like I don't know him. 00:00:53.521 --> 00:00:55.291 Can I call him Jimmy Savage? 00:00:55.296 --> 00:00:56.251 Should I call him lj? 00:00:56.251 --> 00:00:57.601 What, what, what, what was. 00:00:58.066 --> 00:00:59.026 Tell us about Jimmy. 00:01:01.286 --> 00:01:03.086 Don Berry: So nobody called him lj. 00:01:03.116 --> 00:01:04.676 Nobody called him Leonard. 00:01:05.366 --> 00:01:06.806 They would call him Savage. 00:01:07.546 --> 00:01:11.326 Uh, and sometimes with the English word savage, which I'll mention that. 00:01:11.926 --> 00:01:18.676 Um, and, uh, but everybody called him Jimmy, and he wanted to be called Jimmy. 00:01:19.366 --> 00:01:23.806 Uh, the way it happened was serendipitous. 00:01:23.806 --> 00:01:31.036 I mean, the whole story has really, he, um, uh, bad connotations. 00:01:31.036 --> 00:01:33.136 I mean, Jimmy's life was not very. 00:01:34.291 --> 00:01:36.361 Good, especially when he was a child. 00:01:36.361 --> 00:01:42.121 And I'll, I'll get into that because it matters, it really leads to things, uh, in 00:01:42.121 --> 00:01:44.911 his, uh, uh, adult life and his attitude. 00:01:45.241 --> 00:01:47.851 But the big thing with him was his eyesight. 00:01:47.851 --> 00:01:50.581 And he was born with a congenital defect. 00:01:51.511 --> 00:01:55.651 Um, and it was complicated by, 00:01:57.661 --> 00:01:58.831 uh, myopia. 00:01:59.326 --> 00:02:05.386 So that when he read something, he had to take it up to his eyes 00:02:06.136 --> 00:02:11.716 like this, and he turned, and his eyeballs were continually moving. 00:02:12.646 --> 00:02:17.776 Uh, it was a very sad thing that when he was born, back to how he 00:02:17.776 --> 00:02:25.816 got Jimmy when he was born, uh, his mother went through some bad. 00:02:26.101 --> 00:02:30.961 Uh, things associated with the childbirth and wasn't able to pay attention to 00:02:30.961 --> 00:02:32.761 things like, what's the name of the kid? 00:02:33.541 --> 00:02:38.641 Um, and so, uh, at one point a nurse was, you know, didn't have 00:02:38.641 --> 00:02:42.631 a name, so she wrote down a name. 00:02:42.691 --> 00:02:52.831 It was Jimmie with an IE and, um, when it came time to name him, um. 00:02:53.926 --> 00:03:00.736 There was a tentative name, and then the mother sort of picked the name Leonard. 00:03:01.576 --> 00:03:07.006 Uh, and, but the, the nurse kept calling him Jimmie anyway, uh, it 00:03:07.006 --> 00:03:14.746 stuck and he liked it so that he was, it's his middle name, uh, but 00:03:14.746 --> 00:03:21.106 it was also his nickname and, uh, for reasons known only to him, he liked it. 00:03:21.511 --> 00:03:24.421 I, I suppose it was, you know, everybody called him that 00:03:24.421 --> 00:03:26.581 when he was a young person. 00:03:26.581 --> 00:03:28.921 And so he, he went by that name. 00:03:29.491 --> 00:03:36.511 But it was, uh, all his friends definitely called him Jimmy and his 00:03:36.511 --> 00:03:40.351 enemies, uh, probably called him Savage. 00:03:42.526 --> 00:03:47.506 Scott Berry: Okay, so we should let people know how, how, you know, Jimmy, um, uh, 00:03:47.506 --> 00:03:53.776 within this maybe, uh, your relationship and how, how you met him, um, in this. 00:03:55.036 --> 00:03:56.536 Don Berry: So how I met him was easy. 00:03:56.536 --> 00:04:03.736 I was, uh, a student at, at Dartmouth and I didn't know what I was gonna do. 00:04:04.606 --> 00:04:09.046 Um, I had three kids, um. 00:04:09.751 --> 00:04:13.141 I was, uh, a math, uh, math major. 00:04:13.981 --> 00:04:23.221 And, uh, Tom Kurtz, who, uh, I knew from, uh, the, he built along with John 00:04:23.221 --> 00:04:26.431 Kennedy, he built, uh, basic language. 00:04:26.431 --> 00:04:28.381 The basic language, developed it. 00:04:28.411 --> 00:04:33.241 And, um, he, uh, and so he. 00:04:34.066 --> 00:04:39.076 Suggested to me and uh, others in the math department since I was interested in 00:04:39.076 --> 00:04:41.026 probability I should go into statistics. 00:04:41.566 --> 00:04:46.096 So I applied to various places, uh, and I got a fellowship 00:04:46.096 --> 00:04:48.811 at, uh, Yale and, and, uh, 00:04:51.166 --> 00:04:52.156 went to Yale. 00:04:52.156 --> 00:04:55.666 And I met Jimmy when I walked into his office 'cause he was. 00:04:55.936 --> 00:05:00.466 Uh, there, he and Frank Anscombe were the two big names. 00:05:00.496 --> 00:05:02.596 They had recently formed the department. 00:05:02.596 --> 00:05:07.636 Frank was the first chair, uh, and then Jimmy eventually became, 00:05:08.356 --> 00:05:11.086 uh, the chair of the department. 00:05:11.086 --> 00:05:13.156 But it was, uh, it was small. 00:05:14.476 --> 00:05:16.576 Uh, and so I met him. 00:05:16.606 --> 00:05:23.476 Uh, I took a course from him, uh, in the very first semester I was there. 00:05:24.181 --> 00:05:31.951 Uh, it was a course out of feller and it was, um, a, a course that 00:05:31.951 --> 00:05:35.881 undergraduates could take, but it was also for graduate students. 00:05:35.881 --> 00:05:38.311 And so I'll, I'll get into that. 00:05:38.311 --> 00:05:40.021 So anyway, that's how I met him. 00:05:40.336 --> 00:05:40.786 Scott Berry: Okay. 00:05:41.746 --> 00:05:42.256 Awesome. 00:05:42.916 --> 00:05:43.066 Um. 00:05:43.786 --> 00:05:44.926 Did, did he? 00:05:45.091 --> 00:05:45.766 I, I, I. 00:05:45.796 --> 00:05:48.676 He, he became your, your dissertation advisor. 00:05:48.676 --> 00:05:53.656 It'd be interesting to hear about topics and what, what interesting parts of this. 00:05:53.926 --> 00:05:58.936 Did he teach a course using his book, the Foundations of Statistics? 00:05:59.236 --> 00:06:00.676 Did he teach a course in that? 00:06:01.696 --> 00:06:03.586 Don Berry: Uh, maybe he did at some point. 00:06:03.646 --> 00:06:05.656 Maybe he did when he was at Chicago. 00:06:06.016 --> 00:06:08.701 You know, it was published in 1954 and. 00:06:09.301 --> 00:06:10.381 We're talking about. 00:06:10.381 --> 00:06:12.901 I was there in 65. 00:06:13.861 --> 00:06:21.091 Uh, so I, he didn't teach a course in his book, um, but it was sort 00:06:21.091 --> 00:06:27.901 of obligatory, uh, to read his book and to understand what was going on. 00:06:27.901 --> 00:06:34.771 And it was, you know, it was, I, I, I think he's the father of. 00:06:35.521 --> 00:06:42.031 Modern Bayesian statistics, uh, how can you argue about that? 00:06:42.031 --> 00:06:49.291 Obviously, bays and, uh, so bays is, uh, deserves to be called the 00:06:49.291 --> 00:06:51.481 father of, uh, Bayesian statistics. 00:06:51.481 --> 00:06:56.701 But Savage, really, Jimmy really. 00:06:57.751 --> 00:07:00.841 Made it for, uh, bayesians. 00:07:00.871 --> 00:07:06.361 I mean it provided for the first time throughout statistics, it provided for the 00:07:06.361 --> 00:07:15.421 first time a rigorous, uh, definition of what statistics, what statistics was that 00:07:15.421 --> 00:07:17.851 It was in fact a mathematical discipline. 00:07:18.871 --> 00:07:21.031 Uh, and he started out. 00:07:21.451 --> 00:07:23.701 Just a, a little bit of that. 00:07:24.271 --> 00:07:28.981 He started out thinking that statistics, by the way, he was not a statistician. 00:07:29.761 --> 00:07:31.591 He got his degree in mathematics. 00:07:31.591 --> 00:07:36.061 He was working in and wanted to work in physics and chemistry. 00:07:36.451 --> 00:07:38.401 Chemistry turned out. 00:07:39.166 --> 00:07:43.876 You know, that was sort of unfortunate because he couldn't see, uh, and it 00:07:43.876 --> 00:07:45.976 was, uh, how he went into the lab. 00:07:46.396 --> 00:07:53.116 So eventually he really fell in love with statistics and the philosophy of 00:07:53.116 --> 00:07:55.516 statistics, you know, what did data mean? 00:07:56.086 --> 00:07:59.596 And to him, he was. 00:08:02.986 --> 00:08:04.516 He was a polymath. 00:08:04.546 --> 00:08:06.406 He knew everything. 00:08:06.526 --> 00:08:08.746 He knew everything about everything. 00:08:10.246 --> 00:08:14.566 I mean, it's, it is just amazing and I'll give you a few examples of that. 00:08:15.256 --> 00:08:25.966 Um, but he, he knew, uh, about, um, uh, various EE economics, 00:08:25.966 --> 00:08:27.436 you mentioned economics. 00:08:27.856 --> 00:08:28.186 Scott Berry: yep. 00:08:28.246 --> 00:08:33.616 Um, so, so Milton Friedman is, is quoted, and you know, I, we never know if quotes 00:08:33.616 --> 00:08:36.826 are right, but is quoted as saying he's one of the few people I've met. 00:08:36.826 --> 00:08:39.436 I would, he unhesitatingly call a genius. 00:08:40.096 --> 00:08:42.406 Was what Milton Friedman said of Savage. 00:08:42.916 --> 00:08:43.246 Don Berry: Yeah. 00:08:43.246 --> 00:08:48.616 And David Wallace, um, and, you know, everybody that, uh, knew him deeply, 00:08:49.396 --> 00:08:50.956 thought exactly the same thing. 00:08:51.016 --> 00:08:54.646 I mean, it's hard to to know him. 00:08:55.816 --> 00:08:59.536 Without thinking, he has to be the smartest person in the world. 00:08:59.566 --> 00:09:02.776 'cause he seems to know everything and with, with a depth. 00:09:03.196 --> 00:09:04.876 That is phenomenal. 00:09:05.416 --> 00:09:10.126 So, um, so let me give you an example of that. 00:09:11.086 --> 00:09:12.076 And I'll give you an example. 00:09:12.076 --> 00:09:13.936 It's a, it is kind of a trivial one. 00:09:14.656 --> 00:09:16.906 Uh, I mean, I love to work with him. 00:09:16.906 --> 00:09:21.436 We worked on some examples in science that of course he knew about the 00:09:21.436 --> 00:09:23.206 science and he brought me into it. 00:09:23.836 --> 00:09:31.186 Um, he, his childhood had some ops. 00:09:31.216 --> 00:09:32.866 Mostly they were downs. 00:09:33.406 --> 00:09:36.796 Uh, we know about his childhood mostly because of Richard. 00:09:37.456 --> 00:09:41.896 Savage, who was his brother, younger brother by eight years or so. 00:09:42.436 --> 00:09:46.486 Um, and so Jimmy was Richard's father. 00:09:47.296 --> 00:09:57.586 Um, uh, Jimmy's father was instrumental in making Jimmy's life 00:09:57.586 --> 00:10:02.116 good and was a, a, a big positive. 00:10:02.761 --> 00:10:08.041 Um, uh, the rest of his childhood was pretty bad, but, uh, his parents 00:10:08.041 --> 00:10:14.011 bought him an encyclopedia when he was very young, and he read it 00:10:16.981 --> 00:10:19.831 and, and he, he remembered it. 00:10:20.401 --> 00:10:23.161 Uh, just a very simple thing about that. 00:10:23.371 --> 00:10:27.091 Uh, the main thing I want to tell you in a minute. 00:10:28.441 --> 00:10:31.351 Uh, so he one time said to me, he said. 00:10:31.696 --> 00:10:40.966 Uh, Don, uh, uh, Jean, his wife Jean, um, is writing a project on Finland and 00:10:41.656 --> 00:10:45.106 she's looking for celebrities in Finland. 00:10:45.106 --> 00:10:46.816 She can't find many celebrities. 00:10:46.816 --> 00:10:48.586 Do you know any celebrities in Finland? 00:10:49.816 --> 00:10:56.806 Uh, from that they Finn and I said, I think Victor Borger is a, uh, fan. 00:10:57.976 --> 00:11:00.166 And he said, no, Victor Borger is. 00:11:01.096 --> 00:11:01.786 Danish. 00:11:02.596 --> 00:11:03.736 How did he know that? 00:11:03.736 --> 00:11:05.236 Do you know what vi Victor Berger does? 00:11:05.236 --> 00:11:06.166 You probably don't know. 00:11:06.226 --> 00:11:12.166 He was, he was, he was a celebrity. 00:11:12.616 --> 00:11:14.746 Uh, he was a pianist. 00:11:15.436 --> 00:11:17.266 He was a comedian. 00:11:17.776 --> 00:11:24.916 And he had a piano that joked, I mean, he worked with the piano and various jokes in 00:11:24.916 --> 00:11:31.876 the piano, and he was really clever and, uh, talented, uh, but not very well known. 00:11:32.386 --> 00:11:36.796 And he appeared in the TV shows, like, uh, ed Sullivan show some, 00:11:36.826 --> 00:11:39.256 you know, uh, things like that. 00:11:39.706 --> 00:11:44.356 Uh, and, uh, but he wasn't really very well known. 00:11:45.031 --> 00:11:47.401 Then a little bit more known than he is today. 00:11:48.271 --> 00:11:51.991 Uh, so how did he, how did he know that he couldn't watch television? 00:11:52.021 --> 00:11:53.191 I mean, he couldn't see. 00:11:54.181 --> 00:11:59.371 Um, and his, uh, childhood, just, I mentioned Richard. 00:11:59.941 --> 00:12:02.671 Uh, Richard, uh, said, um, 00:12:04.801 --> 00:12:06.871 that, and I wrote this down. 00:12:06.871 --> 00:12:08.281 He was a brilliant child. 00:12:08.626 --> 00:12:12.526 But he paid no attention to what was going on in school because he couldn't 00:12:12.526 --> 00:12:14.266 see what was going on in school. 00:12:15.166 --> 00:12:20.656 Um, and his teachers said, you can't go to college. 00:12:20.656 --> 00:12:24.826 I mean, she put in, they put in really negative things about him. 00:12:25.246 --> 00:12:29.116 According to Richard, uh, the prevailing wisdom in the school 00:12:29.116 --> 00:12:30.826 was that he was feeble-minded. 00:12:31.491 --> 00:12:32.091 Scott Berry: Wow. 00:12:32.506 --> 00:12:35.956 Don Berry: it was, and, and it was a very bad thing for his. 00:12:36.796 --> 00:12:45.886 Um, uh, emotional, uh, circumstance, but so something that involves you, Scott, um, 00:12:48.256 --> 00:12:58.396 mom, your mother, um, was pregnant with you and she would go to the doctor, 00:12:58.606 --> 00:13:00.586 you know, regular visits to the doctor. 00:13:01.126 --> 00:13:02.746 Uh, you have three older brothers. 00:13:03.601 --> 00:13:10.741 Um, and the doctor would inevitably say, you're due for a girl. 00:13:12.781 --> 00:13:15.811 she would tell me that, and I'd say, no, you're not due for a girl. 00:13:16.561 --> 00:13:24.631 Uh, and I, I mumbled things about, you know, uh, um, what the probability was, 00:13:24.661 --> 00:13:29.341 and I was really interested in that separate from my own personal interest. 00:13:29.836 --> 00:13:30.616 You know what? 00:13:31.006 --> 00:13:32.416 How do you do this? 00:13:32.416 --> 00:13:37.516 How do you find the probability that the unborn is, is going to be a boy? 00:13:37.666 --> 00:13:42.226 And this was in the days PR prior to amniocentesis and any of these. 00:13:42.496 --> 00:13:46.006 And it used to be that people would say. 00:13:47.056 --> 00:13:52.426 Um, you know, pregnant woman would come into a room and would, would meet somebody 00:13:52.426 --> 00:13:54.916 and they'd say, oh, you're carrying low. 00:13:55.066 --> 00:13:58.246 It must be a, it must be a boy. 00:13:58.786 --> 00:14:02.896 Uh, and you know, it was really these, uh, sort of old wives 00:14:02.896 --> 00:14:03.706 tales thing. 00:14:04.216 --> 00:14:08.896 But, so I was interested in, in, in, in pressing your mother with 00:14:08.896 --> 00:14:11.866 something that was, uh, uh, legitimate. 00:14:11.866 --> 00:14:12.886 And so I said. 00:14:14.161 --> 00:14:19.411 Uh, to Savage, I said it, it's silly to say, to use a maximum likelihood estimate, 00:14:19.411 --> 00:14:21.451 which is one I know it's not one. 00:14:22.261 --> 00:14:24.151 Uh, so how do I do this? 00:14:24.151 --> 00:14:28.951 How do I find a prior distribution, let's say I'm willing to accept 00:14:29.401 --> 00:14:36.421 that, um, my wife and I have the same probability forever, uh, of, of a boy, 00:14:36.991 --> 00:14:40.381 but it varies potentially between Madrs. 00:14:40.891 --> 00:14:41.491 Um. 00:14:42.076 --> 00:14:45.916 And, uh, so let's assume exchange ability. 00:14:46.846 --> 00:14:54.136 Um, and, uh, but, but how do I find a prior distribution? 00:14:55.156 --> 00:15:01.336 And so he got up and went to his bookcase and pulled out a book by 00:15:01.366 --> 00:15:11.116 Cardo Genie, C-O-R-R-A-D-O-G-I-N-I, who you may know if you're an economist. 00:15:11.641 --> 00:15:18.061 Um, about the Genie coefficient, which is used in, in, in, 00:15:18.151 --> 00:15:19.411 especially in the economics. 00:15:20.101 --> 00:15:24.841 Uh, and Genie turned out to be also a demographer. 00:15:25.801 --> 00:15:31.771 So I looked at the book and in the back of the book in the last half 00:15:31.771 --> 00:15:39.031 of the book or so, was amazing data on families throughout the world. 00:15:39.766 --> 00:15:50.776 A Australia, uh, Africa, the United States on family size and number of boys. 00:15:51.251 --> 00:15:51.471 Scott Berry: Wow. 00:15:52.846 --> 00:15:55.396 Don Berry: And it was truly amazing. 00:15:55.396 --> 00:16:01.246 I mean, when you looked at it, it was absolutely clear that it was not binomial. 00:16:02.026 --> 00:16:03.466 That was absolutely clear. 00:16:04.366 --> 00:16:05.986 Um, and what. 00:16:06.616 --> 00:16:09.496 What what was clear about it was there was a, 00:16:09.521 --> 00:16:09.641 Scott Berry: a, 00:16:10.816 --> 00:16:11.536 Don Berry: um, 00:16:12.346 --> 00:16:14.956 a, more It, it, 00:16:15.286 --> 00:16:16.126 Scott Berry: effects. 00:16:16.306 --> 00:16:20.866 Don Berry: it could have been, uh, it, it, it, it could be, it looked like 00:16:20.866 --> 00:16:30.226 it was a, um, uh, a distribution of, of p and a beta binomial, and, but. 00:16:30.811 --> 00:16:36.451 There was a, an interesting aspect to that, uh, some sort of an 00:16:36.451 --> 00:16:41.761 added effect that if you looked at families with, I mean these number of 00:16:41.761 --> 00:16:45.901 families, number of, uh, of children, the family went up to like 18. 00:16:46.891 --> 00:16:53.971 Uh, there wasn't very much data at 18, but let's say you went up to 12 and, uh, 00:16:54.031 --> 00:16:59.731 the proportion of boys and families of 12. 00:17:01.216 --> 00:17:08.986 Uh, that was one that is, all of them were boys, was bigger than the 00:17:08.986 --> 00:17:12.076 corresponding proportion for 11. 00:17:13.136 --> 00:17:13.426 Scott Berry: Yeah. 00:17:13.546 --> 00:17:18.256 Don Berry: you know that in a binomial it's, it's 12 00:17:18.256 --> 00:17:19.966 times in the other direction, 00:17:20.626 --> 00:17:22.306 so something is going on. 00:17:22.306 --> 00:17:23.716 I eventually learned. 00:17:25.006 --> 00:17:29.266 That, uh, there are some women throughout the world and this was, you know, 00:17:29.266 --> 00:17:31.006 constant, uh, throughout the world. 00:17:31.666 --> 00:17:35.236 Uh, some women who cannot carry a male fetus. 00:17:35.506 --> 00:17:38.116 Uh, and so that would explain a point mass. 00:17:38.116 --> 00:17:43.156 So it was a mixture of a point mass and a beta was fit pretty well. 00:17:43.756 --> 00:17:48.916 And I calculated that the probability you were going to be a boy was 57%. 00:17:49.036 --> 00:17:54.016 Uh, based on that prior probability, of course it went, 00:17:54.016 --> 00:17:56.866 it went up for the next one. 00:17:57.466 --> 00:18:01.096 And, uh, uh, mom and I had two girls. 00:18:01.306 --> 00:18:01.816 Uh, 00:18:03.676 --> 00:18:04.306 So, 00:18:04.531 --> 00:18:08.101 Scott Berry: what what's fantastic about that is, is first of all that somebody, 00:18:08.101 --> 00:18:09.961 he knew exactly what book to go to. 00:18:09.961 --> 00:18:15.001 He remembered that what, what I, what, what's amazing is I, I kind of thought 00:18:15.001 --> 00:18:20.491 maybe he's a pure theoretical guy with the foundations and he came up with this 00:18:20.491 --> 00:18:23.611 amazing axiomatic approach to probability. 00:18:23.901 --> 00:18:26.721 But he worked in economics. 00:18:27.021 --> 00:18:31.371 Apparently he has some, uh, uh, some aspects that maybe he brought 00:18:31.371 --> 00:18:33.411 brownie in motion to asset pricing. 00:18:33.411 --> 00:18:36.351 He worked with Milton Friedman in World War ii. 00:18:36.351 --> 00:18:40.671 He worked with John Von Neuman as a mathematical statistician. 00:18:40.911 --> 00:18:45.501 I remember Edwards Lindman in Savage where it was behavioral sciences 00:18:45.501 --> 00:18:49.521 he worked in, sounds like he was actually an, an applied statistician 00:18:49.521 --> 00:18:51.951 despite being a genius mathematician. 00:18:52.511 --> 00:18:53.231 Don Berry: Yeah. 00:18:53.291 --> 00:19:01.426 And he, he got his, he got his rocks off by looking at, um, uh, 00:19:01.431 --> 00:19:06.191 uh, science and applying statistics. 00:19:06.611 --> 00:19:11.111 I worked with, we used to have, uh, we had a, uh, a, a fish. 00:19:11.896 --> 00:19:18.196 Um, expert that wanted some help and he knew everything about the ocean and about, 00:19:18.196 --> 00:19:22.996 uh, you know, a diurnal effect and, and the like, and was really interested in it. 00:19:23.416 --> 00:19:28.096 We did another thing where I remember it was really neat. 00:19:28.426 --> 00:19:30.406 We did mathematics along the way. 00:19:30.766 --> 00:19:35.836 It turned out that the, uh, the, along the route we saw that, uh. 00:19:36.541 --> 00:19:41.191 Uh, the, what the integral equation was that you had to solve in order to 00:19:41.191 --> 00:19:49.411 do this thing, but it was about a real application, like this thing, um, the 00:19:49.411 --> 00:19:52.681 distribution of of gender in, in families. 00:19:53.746 --> 00:19:54.136 Scott Berry: Yeah. 00:19:54.556 --> 00:19:59.506 So I, it it, was he interested in clinical trials? 00:19:59.506 --> 00:20:03.226 I mean, you went into bandit problems, you talked about the 00:20:03.226 --> 00:20:04.846 bandit problems in clinical trials. 00:20:04.906 --> 00:20:07.756 Was he interested in clinical trials and how did you, how did 00:20:07.756 --> 00:20:09.256 you come to the topic with him? 00:20:11.236 --> 00:20:13.501 Don Berry: So, um, 00:20:16.041 --> 00:20:17.626 it, it was my topic. 00:20:17.926 --> 00:20:19.756 Uh, he was not. 00:20:20.401 --> 00:20:21.901 Into clinical trials. 00:20:21.961 --> 00:20:24.271 I mean, he was sort of like fisher in that way. 00:20:24.271 --> 00:20:28.621 Fisher was amazing with his, uh, agriculture stuff. 00:20:29.281 --> 00:20:32.581 Um, he was not into clinical trials. 00:20:33.091 --> 00:20:37.381 He, he became interested in clinical trials because of me. 00:20:38.281 --> 00:20:41.281 I mean, this, he was interested in strategy. 00:20:41.281 --> 00:20:46.621 You know what, uh, his other most famous book is, uh, how to Gamble if you Must. 00:20:47.356 --> 00:20:51.286 Where, uh, it was based on a utility function that's 00:20:51.286 --> 00:20:55.246 different than a typical, um. 00:20:55.936 --> 00:20:59.416 Namely, there's an amount of money that you really want and need 00:21:00.316 --> 00:21:04.696 to buy an airplane ticket to go home and nothing else matters. 00:21:05.026 --> 00:21:08.116 And so it's utility, one above that and utility. 00:21:08.416 --> 00:21:14.446 And then there are some strange things that happen, uh, in the strategy and 00:21:14.446 --> 00:21:17.146 in particular, bold play is optimal. 00:21:17.716 --> 00:21:23.866 Um, and so he took this seemingly small problem. 00:21:24.421 --> 00:21:28.081 And, uh, built finite adaptivity. 00:21:28.171 --> 00:21:33.181 It made it show why you need finite adaptivity and why it's more important. 00:21:33.901 --> 00:21:40.261 Um, and so it was, and, and that was not unusual, where at the end of the road 00:21:40.621 --> 00:21:43.111 was a revelation about the mathematics. 00:21:43.981 --> 00:21:52.621 Uh, so, uh, anyway, he was, he was, um, not interested in clinical trials. 00:21:53.041 --> 00:21:57.121 The one person who was was Frank anco. 00:21:58.006 --> 00:21:58.306 Scott Berry: Ah. 00:21:58.531 --> 00:22:06.001 Don Berry: so my initial, when I, where did the Bandit problem come from? 00:22:06.601 --> 00:22:13.621 From my perspective, I took a course in the management, uh, uh, department. 00:22:13.621 --> 00:22:15.361 It was like the business school at Yale. 00:22:16.111 --> 00:22:20.941 Uh, and there, uh, I came across a problem. 00:22:20.941 --> 00:22:21.811 It was kinda like. 00:22:22.501 --> 00:22:28.501 The one arm bandit where you, uh, ask yourself, should I keep playing this game? 00:22:28.921 --> 00:22:30.781 Should I keep pulling the one arm? 00:22:31.321 --> 00:22:41.701 Um, and so, uh, you, but you learn over time and you learn, uh, you know, when 00:22:41.701 --> 00:22:46.471 you've, when you've had enough and you're convinced that you're never gonna win 00:22:46.471 --> 00:22:49.231 at this, so you go to do something else. 00:22:49.621 --> 00:22:51.421 Uh, and so I. 00:22:51.481 --> 00:22:56.071 I said, you know, the two armed version is even more interesting. 00:22:56.791 --> 00:23:02.551 And so I started working on it, did some dynamic programming, 00:23:02.551 --> 00:23:07.171 did some computations, and I really became, uh, interested. 00:23:07.921 --> 00:23:12.061 Um, and so I got Savage interested 00:23:12.751 --> 00:23:15.571 Scott Berry: Did, did you think about ANCO as an advisor? 00:23:16.711 --> 00:23:17.761 Don Berry: I did. 00:23:18.661 --> 00:23:24.601 And I, I, you know, I, I, I thought he was, uh, an amazing mind. 00:23:24.661 --> 00:23:32.611 I thought he was, um, a, he was an amazing mind, and a, and a great person, and 00:23:32.611 --> 00:23:35.431 a, and a wonderful person to work with. 00:23:35.881 --> 00:23:37.261 But I was, so, 00:23:39.901 --> 00:23:44.371 so one of the things I said, uh, in my comments about, 00:23:44.491 --> 00:23:47.581 uh, Savage, it, it, I said. 00:23:48.286 --> 00:23:52.486 It was like the world around you. 00:23:52.486 --> 00:24:00.466 When you're with Savage, the world around you is tingling with intellect and it, 00:24:00.526 --> 00:24:05.236 it, it, it, that was so attractive. 00:24:05.926 --> 00:24:10.336 And so, I mean, I just, and I knew he was interested in strategy 00:24:11.056 --> 00:24:13.726 and I knew that this problem would really interest him. 00:24:13.726 --> 00:24:15.766 I didn't know the, the name Bandit. 00:24:16.096 --> 00:24:18.346 Eventually I knew that it was called Bandits. 00:24:18.916 --> 00:24:20.326 It's a very hard problem. 00:24:20.656 --> 00:24:27.016 Uh, suffice to say that the computer scientists call it NP-hard Uh, suffice to 00:24:27.016 --> 00:24:29.836 say that, uh, Peter Whittle, who was uh. 00:24:30.391 --> 00:24:33.691 Uh, uh, worked in World War ii. 00:24:34.081 --> 00:24:39.901 It was a, a statistician, uh, worked in World War II on, uh, strategic things. 00:24:40.261 --> 00:24:52.621 Said that, um, they talked in, in the UK about out making a one 00:24:52.621 --> 00:24:57.961 sheet description of the bandit problem and dropping it on Germany. 00:24:58.711 --> 00:25:02.971 So that the mathematicians would get hold of it and waste their 00:25:02.971 --> 00:25:05.611 time trying to solve this problem. 00:25:06.316 --> 00:25:06.736 Scott Berry: Uh, 00:25:06.841 --> 00:25:09.151 Don Berry: Uh, so it was really hard. 00:25:09.241 --> 00:25:14.641 Uh, but it was, so I thought it was, I was doing clinical trials 00:25:15.301 --> 00:25:15.591 Scott Berry: yeah. 00:25:16.171 --> 00:25:19.951 Don Berry: and um, of course I wasn't. 00:25:20.131 --> 00:25:23.491 And that was the, you know, that's the rest of my story. 00:25:24.571 --> 00:25:25.201 Um. 00:25:26.476 --> 00:25:29.866 So the, what else did I want to 00:25:30.316 --> 00:25:34.816 Scott Berry: So, so, uh, that, interestingly, it ties into the, my 00:25:34.816 --> 00:25:37.696 comment earlier of, uh, academic brothers. 00:25:37.696 --> 00:25:43.216 So, Jay Cade was part of your committee, uh, within this Jay Cade. 00:25:43.276 --> 00:25:47.146 Uh, interestingly, Moy DeGroot was a student of his, and Jay 00:25:47.146 --> 00:25:48.916 Cade went to Carnegie Mellon. 00:25:49.381 --> 00:25:53.431 Uh, I went to Carnegie Mellon and became a student, uh, of j Cade, 00:25:53.431 --> 00:25:55.171 and he was my thesis advisor. 00:25:55.441 --> 00:26:01.021 So I I, in some ways I'm almost your academic brother in this and, um, my 00:26:01.021 --> 00:26:07.231 daughter Lindsay, um, interestingly her last two choices were CMU and Duke. 00:26:07.261 --> 00:26:11.881 And if had she gone to CMU, she may have been j Kade student, and so it, it 00:26:11.881 --> 00:26:13.651 may have been that much more twisted. 00:26:13.651 --> 00:26:15.271 And interestingly, Lindsay. 00:26:15.841 --> 00:26:19.651 Was an honorable mention for the LJ Savage Award for her 00:26:19.651 --> 00:26:22.021 dissertation In Bayesian statistics. 00:26:23.081 --> 00:26:23.771 Don Berry: It's wonderful. 00:26:24.151 --> 00:26:24.571 Scott Berry: Yeah. 00:26:24.886 --> 00:26:27.406 Don Berry: So I, I, I have to mention, um. 00:26:28.066 --> 00:26:32.446 The dissertation and, uh, Savage's role in it. 00:26:33.496 --> 00:26:37.396 He was, I showed him my first draft. 00:26:37.936 --> 00:26:41.836 My first draft had only the beta distribution and I had some really 00:26:41.836 --> 00:26:46.786 nice theorems, uh, when the prior distribution is the beta distribution. 00:26:47.686 --> 00:26:49.306 It was 10 pages long. 00:26:50.266 --> 00:26:51.886 And he said, well, this is fine. 00:26:51.946 --> 00:26:55.246 This is a dissertation, but let's try a little bit more. 00:26:56.086 --> 00:26:59.836 And let's try to, you know, understand more generally. 00:27:00.316 --> 00:27:05.806 And so it ended up going through five drafts and I tell you that every single 00:27:05.806 --> 00:27:11.626 one of them, he looked at every single word he recorded what he was doing. 00:27:11.626 --> 00:27:15.076 You have to understand it's a little bit delicate 'cause you 00:27:15.076 --> 00:27:16.786 know, of his writing and, and. 00:27:17.281 --> 00:27:18.121 Uh, eyesight. 00:27:18.781 --> 00:27:26.041 Uh, but he would record his, what he was reading and explaining to me. 00:27:26.161 --> 00:27:27.811 He taught me how to write. 00:27:28.111 --> 00:27:28.771 Scott Berry: Wow. 00:27:29.341 --> 00:27:35.791 Don Berry: Um, and it, it, uh, it, it, it sort of carried over. 00:27:35.791 --> 00:27:39.061 I mean, you've learned a little bit of things from me for writing, 00:27:39.511 --> 00:27:41.491 and it's, it comes from Savage. 00:27:42.001 --> 00:27:46.741 Um, it was, the end was so well written. 00:27:47.506 --> 00:27:54.316 That when I submitted it to the Annals of Mathematical Statistics, it was 00:27:54.496 --> 00:27:58.186 published in the last year of the Annals of Mathematical Statistics. 00:27:58.186 --> 00:28:04.456 After that, it split into two Annals of Statistics, annals of Probability, uh, and 00:28:04.456 --> 00:28:10.756 it, the associate editor was Tom Ferguson, a very famous decision theorist, and 00:28:11.386 --> 00:28:14.896 Ferguson didn't send it out for review. 00:28:16.366 --> 00:28:21.946 He read it himself, he approved it himself. 00:28:22.996 --> 00:28:31.276 It was published 27 pages and there were no revisions at the journal level. 00:28:32.326 --> 00:28:38.536 Um, and it was J it was, it was impeccable because of Jimmy who was. 00:28:38.581 --> 00:28:41.341 Uh, uh, he was a perfectionist. 00:28:41.341 --> 00:28:42.931 It was his downfall. 00:28:42.931 --> 00:28:47.761 Some you mentioned, uh, some other people that might not have been. 00:28:47.761 --> 00:28:49.741 So, I mean, I loved Savage. 00:28:49.801 --> 00:28:51.661 I mean, he was like a father figure. 00:28:51.661 --> 00:28:53.551 He was a father figure for me. 00:28:54.301 --> 00:28:56.521 Um, and I'd have done anything for him. 00:28:57.091 --> 00:29:04.561 Um, uh, but other people had been in conversations with him that. 00:29:05.176 --> 00:29:07.396 Uh, we're not all that positive. 00:29:07.396 --> 00:29:11.056 I know Herbert s Shernoff, who I think is an amazing person. 00:29:11.056 --> 00:29:12.286 I, and I love him too. 00:29:12.856 --> 00:29:13.726 He's still alive. 00:29:13.726 --> 00:29:15.376 He's 101. 00:29:16.246 --> 00:29:25.006 Um, and, uh, so I, I said to him he's interested in, in sequential things. 00:29:25.786 --> 00:29:28.696 Uh, he was Jake Kaine's advisor, 00:29:28.861 --> 00:29:29.401 Scott Berry: Yeah. 00:29:29.491 --> 00:29:29.911 Yeah. 00:29:30.016 --> 00:29:32.956 Don Berry: um, uh, and, and. 00:29:33.301 --> 00:29:37.891 Uh, so I, I was ex we were talking about things, mutual interest 00:29:37.891 --> 00:29:41.011 and, and sequential things, and he was interested in bandits. 00:29:41.221 --> 00:29:45.181 I don't, uh, turn off that is, I don't think he ever published on Bandits, 00:29:45.181 --> 00:29:49.471 but he was very interested in them and he had done some calculations on them. 00:29:49.921 --> 00:29:52.471 So the subject came up of Jimmy Savage. 00:29:53.776 --> 00:29:58.366 And so I mentioned, you know, something about Savage and he said some very 00:29:58.366 --> 00:30:01.246 disparaging things about Savage. 00:30:01.246 --> 00:30:13.426 He and I said, well, but Herman, he, he was human and Chernoff said 00:30:14.866 --> 00:30:17.776 he had some human characteristics. 00:30:20.161 --> 00:30:25.201 So it, it, it, and, and, you know, everything comes back to the. 00:30:25.831 --> 00:30:28.981 The, the eyesight and the way he was treated. 00:30:29.491 --> 00:30:33.661 Uh, let me just mention, and I said I was gonna say something about Bill Cleveland. 00:30:34.141 --> 00:30:36.631 Uh, Janice Cleveland was his wife. 00:30:36.631 --> 00:30:40.501 We had a party one night at, at, at, uh. 00:30:41.171 --> 00:30:47.111 Uh, Savage's house and, uh, there were the, the, the women, the, the 00:30:47.111 --> 00:30:50.351 wives mostly, uh, were talking. 00:30:50.921 --> 00:30:58.931 Um, and Jimmy came up and, uh, entered the conversation and, uh, uh, Janice 00:30:58.931 --> 00:31:04.001 Cleveland said, uh, uh, professor Savage we're talking about children and, you 00:31:04.001 --> 00:31:05.981 know, what are the good ages for children? 00:31:05.981 --> 00:31:08.411 What's your favorite age for children? 00:31:10.921 --> 00:31:15.331 And he said, your age. 00:31:17.221 --> 00:31:18.781 And now people laughed. 00:31:19.351 --> 00:31:28.291 But it was real from his, it was from, he was serious from his experience. 00:31:28.891 --> 00:31:32.341 Uh, he didn't want to be exposed to these people that had really 00:31:32.341 --> 00:31:34.861 done some, uh, nasty things. 00:31:34.861 --> 00:31:37.801 I mean, you can imagine the bullying and the like. 00:31:38.821 --> 00:31:42.391 Uh, so it's a Sergeant Freud. 00:31:43.171 --> 00:31:47.641 It was a, a very good story. 00:31:48.271 --> 00:31:49.891 Uh, he died too soon. 00:31:49.891 --> 00:31:52.381 He died at age 53 from angina. 00:31:52.951 --> 00:31:56.851 Scott Berry: Yeah, it was, it's incredible all he accomplished to work in the war 00:31:56.851 --> 00:32:03.511 and economics and the, the, the two books, uh, the students he had and he was 53. 00:32:03.751 --> 00:32:06.811 Uh, it's unbelievable how much he accomplished in that time. 00:32:07.741 --> 00:32:08.011 Don Berry: yeah. 00:32:08.826 --> 00:32:09.116 Scott Berry: Yeah. 00:32:09.316 --> 00:32:11.266 Don Berry: I, I, I haven't told you. 00:32:11.266 --> 00:32:12.556 Maybe we can do another one. 00:32:13.336 --> 00:32:19.456 Uh, can, can I tell you about, um, my experience with, uh, 00:32:19.516 --> 00:32:22.096 teaching and, and, and with Savage? 00:32:22.696 --> 00:32:23.236 Uh, 00:32:23.686 --> 00:32:24.916 Scott Berry: let's end it with that. 00:32:24.916 --> 00:32:26.116 So make it a bang. 00:32:26.146 --> 00:32:27.196 Make it a bang. 00:32:27.796 --> 00:32:28.186 Don Berry: Alright. 00:32:28.186 --> 00:32:28.936 Here's a bang. 00:32:30.526 --> 00:32:34.906 So, uh, I had experience with. 00:32:35.461 --> 00:32:40.051 With, uh, Jimmy in his class, we, I, I took where he said, 00:32:40.051 --> 00:32:43.501 if ever I don't show up. 00:32:44.071 --> 00:32:44.731 To class. 00:32:44.731 --> 00:32:50.461 He told me, he said, you take the class and, and uh, just, uh, 00:32:50.671 --> 00:32:54.211 you know, uh, continue to where you think we should be going. 00:32:54.961 --> 00:32:59.611 So I was hoping and hoping that wouldn't happen, but one day it did happen. 00:33:00.691 --> 00:33:06.421 Uh, and so I got up and I said to the class, which was mostly 00:33:06.421 --> 00:33:08.881 undergraduate math majors at, uh, Yale. 00:33:09.841 --> 00:33:11.881 Uh, so we had homework last night. 00:33:11.881 --> 00:33:13.651 Anybody have questions about the homework? 00:33:14.566 --> 00:33:19.036 And, uh, there was a question and the guy asked, uh, really the 00:33:19.036 --> 00:33:21.856 hardest question in, in the homework. 00:33:22.486 --> 00:33:28.606 Uh, and it had to do with the acidotic, uh, tail of the normal distribution. 00:33:29.536 --> 00:33:32.896 And so I showed him how to do it. 00:33:33.376 --> 00:33:39.886 I started to show him how to do it, and then Savage came in and, um. 00:33:41.311 --> 00:33:43.231 I went up and I handed the chalk to him. 00:33:43.231 --> 00:33:44.521 He said, no, you continue. 00:33:45.961 --> 00:33:52.891 So I continued and uh, at, I showed him how, how to do the problem, but 00:33:52.891 --> 00:33:59.521 then I related the problem to some real things and this impressed Savage. 00:33:59.671 --> 00:34:07.951 Um, uh, and he then asked me, uh, if I would be willing to teach a course. 00:34:08.671 --> 00:34:14.281 Uh, at Albertas Magnus College, a Albertas Magnus College was in New Haven. 00:34:14.341 --> 00:34:15.241 Still is. 00:34:15.271 --> 00:34:18.811 It was, uh, all girls co-ed now. 00:34:19.471 --> 00:34:24.601 Uh, but they had a statistics course that had not, um, 00:34:24.606 --> 00:34:24.826 Scott Berry: um, 00:34:26.311 --> 00:34:31.321 Don Berry: uh, they lost the teacher and so they asked Savage if he could 00:34:31.321 --> 00:34:32.851 provide somebody that would do this. 00:34:32.851 --> 00:34:37.621 And so he asked me if I would do it, and he said it's $400. 00:34:37.996 --> 00:34:39.946 This is a whole semester for $400. 00:34:40.516 --> 00:34:43.576 Uh, uh, and I said, sure, of course. 00:34:43.576 --> 00:34:45.886 I, and, you know, in part because of Savage. 00:34:46.636 --> 00:34:54.826 Um, and then I bought a car, um, a 58 Chevrolet with the money. 00:34:55.336 --> 00:35:01.576 And, um, I would give Savage a ride home, uh, with, with the car. 00:35:01.936 --> 00:35:06.166 And so one night he got into the car and he said, so how's Big Al? 00:35:08.116 --> 00:35:11.176 I said Big L. 00:35:11.866 --> 00:35:12.766 Who's Big L? 00:35:13.696 --> 00:35:16.546 And he said, Albertas Magnus. 00:35:16.546 --> 00:35:18.856 The car, the car's name is Big L. 00:35:20.476 --> 00:35:23.986 And so that's part of the story. 00:35:23.986 --> 00:35:27.796 But he, and then he kept calling it Big L but then we had a very 00:35:27.796 --> 00:35:32.896 famous guy, and you've probably heard of him, um, Fred Mueller. 00:35:34.321 --> 00:35:34.921 Scott Berry: Oh yeah, 00:35:34.996 --> 00:35:38.836 Don Berry: Uh, I think it was Feinberg's, uh, uh, advisor, 00:35:39.826 --> 00:35:41.206 uh, very famous guy. 00:35:41.206 --> 00:35:44.686 Bayesian didn't write a lot about Bayesian. 00:35:44.686 --> 00:35:48.106 He did the Bayesian thing of the Federalist, Federalist papers. 00:35:48.106 --> 00:35:52.606 You know, the authorship, the Federalist paper, uh, uh, that was 00:35:52.606 --> 00:35:55.876 Bayesian and that was, uh, him and, uh. 00:35:57.226 --> 00:35:58.606 Uh, somebody else. 00:35:58.696 --> 00:36:09.616 Uh, anyway, so Fred Moeller visited and, uh, lots of people visited, uh, 00:36:09.646 --> 00:36:12.436 new Haven when, uh, Savage was there. 00:36:12.466 --> 00:36:16.931 You know, Dennis Linley, um, George Barnard. 00:36:17.911 --> 00:36:24.511 Uh, uh, uh, Jerry Kornfield, I mean, it was a mecca to go to. 00:36:25.081 --> 00:36:31.171 Um, and so, uh, when he was going back, Savage asked me if I would take 00:36:31.171 --> 00:36:34.861 him back to the airport, and he wanted to go along with, because they had 00:36:34.861 --> 00:36:37.171 worked together, he and, and Ello. 00:36:38.491 --> 00:36:39.271 So the. 00:36:39.406 --> 00:36:42.376 Scott Berry: in the car with Jimmy Savage and Fred Mosell is 00:36:42.526 --> 00:36:43.696 they're, they're, they're chatting. 00:36:43.696 --> 00:36:44.476 That's awesome. 00:36:44.806 --> 00:36:47.176 Don Berry: It's chit chatting and they started to talk 00:36:47.176 --> 00:36:49.756 about some scientific problem. 00:36:51.436 --> 00:36:56.566 And the, i, I, I was lucky on that, uh, thing where I talked 00:36:56.566 --> 00:37:00.196 about the asymptotic distribution, the asymptotics of the normal, 00:37:00.766 --> 00:37:01.246 uh, I. 00:37:01.306 --> 00:37:05.716 Scott Berry: Mill, Mills ratio I'm sure came into that, but, but, but, but, yep. 00:37:05.926 --> 00:37:07.726 Don Berry: but here I was lucky too. 00:37:08.356 --> 00:37:13.576 Uh, they were talking about the scientific thing and I kept, uh, 00:37:13.906 --> 00:37:18.886 adding things and commenting on what the scientific problem was. 00:37:19.036 --> 00:37:26.266 And so, uh, Jimmy says to me, Don, how do you know so much about this? 00:37:27.856 --> 00:37:35.446 And I said, well, I, I read the article in the science encyclopedia 00:37:35.446 --> 00:37:41.956 in the stat library, and his eyes lit up and he was delighted. 00:37:41.956 --> 00:37:44.356 He said, you know, I, I worked so hard. 00:37:44.476 --> 00:37:45.646 I had to convince. 00:37:45.961 --> 00:37:49.261 Yale to let me buy those, that thing. 00:37:49.441 --> 00:37:53.101 And so what you've said now makes it all worthwhile. 00:37:53.131 --> 00:37:58.021 That it is a useful thing and it, it hearkens back to the 00:37:58.441 --> 00:38:00.481 encyclopedia that he read. 00:38:00.541 --> 00:38:02.611 You know, I mean, he was really. 00:38:02.971 --> 00:38:08.731 A human encyclopedia, but that's too technical a thing. 00:38:08.731 --> 00:38:10.771 I mean, it isn't, he wasn't technical. 00:38:10.891 --> 00:38:20.611 He was brilliant and innovative and, uh, uh, uh, I'm, I'm, I'm, I'm glad 00:38:20.611 --> 00:38:22.711 we had him in the Bayesian world. 00:38:22.891 --> 00:38:26.671 We probably wouldn't exist, Scott, if it were not for him. 00:38:27.131 --> 00:38:30.961 Scott Berry: So Berry consultants, uh, has a lot to thank for, 00:38:31.051 --> 00:38:32.671 for, uh, Jimmie Savage. 00:38:32.671 --> 00:38:38.641 And so let's end it with, uh, three statisticians and a 58 Chevy named Big Al. 00:38:39.601 --> 00:38:43.771 Alright, so, so Dad, thanks a lot. 00:38:43.801 --> 00:38:44.941 Appreciate it. 00:38:45.181 --> 00:38:46.951 Until next time, in the interim. 00:38:47.291 --> 00:38:47.721 Don Berry: Thank you. 00:38:48.496 --> 00:38:49.486 Thanks everybody. 00:38:49.816 --> 00:38:50.236 Bye.