In today’s episode, Dan Murphy explains how to use exponential growth portfolios to build a once-a-month training strategy using the Boss AI.
Dan opens the show by drawing evocative parallels between maximizing the potential of AI for trading, and Mao’s war on sparrows during the 1958 Great Leap Forward amid the Communist uprising in China. Just as Mao’s top-down approach led to the deaths of millions, Dan makes the point that overspecifying what a computer—in this case, the Boss AI—should look for begets the dangers of “cracking the whip” as opposed to “dangling the carrot”. After all, Dan adds, drawing another, closer parallel, capitalism is at its finest when all parties are given the freedom to be creative from a bottom-up approach.
Listen in as Dan discusses factors in evaluating the fitness of a trading strategy, capitalizing on exponential growth stocks via DB Transactions, why you should look at stock trading like you would sports betting, and the importance of being able to articulate fitness functions and your chances of making a good trade.
- What using AI for trading and Mao’s Great Leap Forward have in common
- The big blind spot in conventional thinking around optimizing trading strategies with the Boss AI
- A tutorial on navigating DB Transactions portfolios consisting of exponential growth stocks
- Minimizing top-down direction to give the Boss greater flexibility and allow it to do the heavy lifting
- Getting scientific with evaluating your odds of success and failure when trading
- Using the Boss to create a winning, long-term strategy based around your goals and circumstances
Dan: Hey Dan Murphy here and today we have a very exciting show. We are going to use exponential growth portfolios to build a once a month, just once a month trading strategy from scratch using the Boss AI. Since this is an Ai show, we're going to go ahead and use portfolios that are 3D printing, alternative energy cloud computing cyber security driverless revolution. Exponential growth stocks in general the Nasdaq 100, which is tech-laden, nanotech stocks, robotics social media. It's going to be a whole lot of fun so stay tuned.
INTRO: Nowadays, computers do everything. So, why shouldn't we use computers to make money? This is artificial intelligence real profits with Dan Murphy. Learn about automatically building profitable trading strategies using Ai and a supercomputer. Build tailor-made super ai trading strategies from scratch. Without writing a single line of code, get ready to have your mind blown. And now your host from beautiful Newport coast in California get ready for no BS, no-nonsense, no news-driven trading with your host Dan Murphy.
Dan: Let's start up today's show with Mao’s war on sparrows. And of course, this is 1958, Mao had just had the communist uprising. China had been humiliated in what they call the century of humiliation where they had lost the Opium Wars to the west and they were trying to rebuild the country. Unfortunately, of course, they got stuck in communism. So, one of the things that Mao decided to do in his great leap forward was killing off a bunch of pests. Because they're having a pestilence problem starvation problem of course, which tends to happen in top-down economies. We saw that USSR as well.
So, Mao had a war against what he called ‘The Four Pests’. And so that would be mosquitoes, which makes sense they carry a whole lot of disease. One that's the deadliest animal in the world right. Rats, okay they carry bubonic plague. Which I believe the bubonic plague did originate in china before it moved over to Europe and killed what half the population. Flies, yeah you can see a benefit of flies I suppose and sparrows. Sparrows, they all had to be eliminated according to Mao. Basically, his thinking and rationale was that, “Hey, sparrows must be eating lots of grain and therefore taking away and depriving his workers from food.”
Well, they had great success eliminating the sparrows, you know just in Shanghai they use some just really primitive weapons to annihilate. Almost 200,000 sparrows, the communists are always really good at eliminating and killing things for sure especially people. And they even celebrated a 16-year-old boy, named Yang Shimon. He was considered a national hero because he basically found the nest of all these sparrows and personally murdered with his bare hands about 20,000, which might be exaggerated.
But in 1959, the academy of sciences in china well they brought something to Mao’s attention. That was the fact that, when they dissected these sparrows, they found out that the digestive tract contained about three-quarters harmful insects. Like let's say grasshoppers or locusts that were eating the food and then just a little bit of grain in there. So, what they found was that sparrows were actually helping people eliminate these green eating, food eating pests. And what do you know, there was a famine that killed about 30 million people and it looks like Mao’s top-down approach helped further along their demise.
It just goes to show you about top-down economies that you have some fool at the top that has no idea what he's talking about. And all these millions of people listening to that person, because they have tremendous influence and if you didn't listen you probably were murdered right. What does that have to do with trading? Well, one of the things that from an evolutionary standpoint that the Boss Super Ai does is, it starts at the very bottom and just does minimal improvements over time every generation.
So, you start with all these random strategies. There's no top down direction other than a very broad stroke. Let's say I want to make 50% a year company on the growth rate or 80% a year companional growth rate. I want to have low drawdowns let's say less than 30%. I want to trade five days on average. The average length of my holding period is five days. If you want to do short term or let's say 20 days, so that would be about a month. The same thing could be said in a capitalist economy that you wouldn't want to have too much direction. You want to go ahead and dangle a carrot for individuals, so that they can improve and see what they want to do as and do what they want to do as they see fit.
So, one of the ways you could do that in a capitalist society is tax incentives. You just don't tell people exactly what to do you just give them a generality, which is exactly what we're doing with the Ai. So, we want to go ahead and articulate certain things. Like maybe I want every single year 100% of years to be winning years. That could be something you can articulate in what's called ‘A Fitness Function’, for designing these trading strategies. But at the same time, you're just keeping in mind that you are not telling the computer exactly what to do from the very top.
I think that that's something that perhaps we could be getting in trouble with right now as the west has completely forgotten about Malice China and forgotten about USSR. And how those systems did not work, because they cracked the whip instead of dangling the carrot. Now when it comes to designing these trading models, what they do is they start very randomly in the first generation. And let's say that we have a population size of 100, trading strategies is randomly thrown together. Then basically you just start breeding the strategies that are making just a little bit more than the others and then you kind of kill off those that are making less. Then each generation you see these little incremental improvements. Until let's say at generation 100
All of a sudden, you just see these amazing trading strategies coming out of nowhere. And sometimes, there's sudden jumps in the performance that there's something some kind of random mutation that happens. Just like, let's say a peacock. That was always Darwin’s dilemma when he looked at peacocks and said, “Well, how do they even exist? How does a peacock exist? Because it doesn't make any sense.” The reason it doesn't make any sense is because so much energy seems to be wasted on the aesthetics of a peacock. But what they found in later computer simulations years after his death was that, these peacocks would attract stronger and stronger mates.
And eventually we would have a bird population that completely branched off all on its own creating a whole new subspecies, which we now see as the peacock. A tremendous amount of energy is spent on let's say antlers as well. But at the end of the day of the day these things tend to help considerably in the survival of the fittest of these animals. And the same thing often happens in trading strategies as well. All of a sudden, you'll see this Cambrian explosion where the fitness just goes up a huge amount. Fitness of course is combining all those traits that I talked about.
So, combining company under growth rate, combining your max drawdown. You know how often you trade? How many winning, what's the percentage of winners? How many percentage winning years are there? Let's say 100&. If you combine all those scores together that will be your fitness and the Boss Super Ai will do its damnedest to go ahead and find that strategy that matches you. Because if you can't trade the strategy what good is it, right. So, I think that that's a big blind spot in a lot of people's thinking. Oh, and in case you're wondering, what did Mao do after he learned that sparrows were actually helping people?
Well, he just quietly took them off the list and then replace them with bed bugs. Hey, what can I say? Crazy commies, right. By the way, I’ve written two reports that you can go ahead and download for free. One on the commitment of trader’s report which is called, ‘The Ultimate Crash Detector’, and another called, ‘The Relaxed Investor’. Even another surprise bonus here, artificial intelligence real profits where you can learn more about the supercomputer. All you need to do is mosey on over to ‘Portfolioboss.com’, and just click on the little free report’s menu. That is ‘Portfolioboss.com/freereports.’
It's up at the top, go ahead and download all three. Actually, there's a few more over there too if you want to go ahead and binge read.
INTRO: This is artificial intelligence real profits with Dan Murphy.
Dan: Okay, so inside of Portfolio Boss, we have a bunch of what I call ‘Power Templates’. Which are basically templates for trading everything from the SOP 500, QQQ, ETS the entire Nasdaq 100 and SOP 500, gold and silver switching. Today we're going to talk about what I call the ‘DB Transactions Portfolios’, which consist of exponential growth stocks. Like driverless revolution, social media stocks, nano technology, the entire Nasdaq 100. Just because it's so tech laden and on and on and on solar energy that and so forth. Because those are a lot of the really disruptive companies that are out there.
That as you can see with amazon, they have been taking over brick and mortar left and right especially gaining in their power with the Coronavirus as many large companies are doing. What you can do is, go ahead and follow along. If you go to ‘Portfolioboss.com’, you go to ‘Podcasts’ and then you can go ahead and see a video of what I'm doing here. So, I’m just going to double click on the Boss template here for DB Transactions. Literally, the only thing I have to do is enable the divine engine and I push ‘Start the Boss’ and click ‘Okay’.
Then I’m immediately running the Boss to discover trading strategies from scratch. Using a once a month in this case, periodically switching positions on a certain trading day of the month. When it's not in stocks, it'll go ahead and get into long-term bonds in this particular example. So, it's basically testing since 2002 and of course you could go ahead and instead of using long-term bonds, you could go ahead and use short-term bonds. You could test this strategy all the way back to the 80s. You could put a symbol for the cash equivalent, as something like PTSHX which is a PIMCO fund.
I actually just moved offices by the way and my new office is basically right next door to PIMCO, Small World. I suppose I've moved it over to fashion island in Newport Beach. But anyway, this particular trading strategy template is using 10 total positions to hold and it's buying the cream of the crop. Now one of the biggest edges of course in what we do with Portfolio Boss is we have a ranking algorithm. So, we want to go ahead and instead of let's say you're trading the Nasdaq 100, why would you want to go ahead and create a trading strategy on every single one of those stocks. Why not just trade the top five or top 10 or 15 stocks at once? And forget all the rest, right.
So, just throw away the rest and just trade the top ones. It makes for a great trading strategy. So, the Boss easily finds trading strategies with these groups of stocks. So, again 3D printing, alternative energy, cloud computing, cyber security driver’s revolution, nanotech robotics and social media to just name a few. Then I'm just going to go ahead and wait for what the boss will discover. In this particular case, I went ahead and made sure just to have a currency volume filter. Where I'm saying, I don't want the average traded volume any lower than 1 million. So, I want to go ahead and make sure I’m trading liquid stocks not penny stocks.
In this particular case too, it was interesting that there is a bi-filter that the Boss discovered, which was basically a volatility range percentile filter. I know that's kind of a mouthful. What it discovered is, it wanted to go ahead and make sure that we go ahead. And if the volatility is less than the 95th percentile, I can go ahead and trade it. So basically, it's filtering out the top 5% most volatile stocks from that entire list and not trading it. Because it's just going to swing your portfolio around wildly if you're trading the most volatile of the stocks. So, pretty cool how it actually filtered those out at the top end
A lot of times when people are trading momentum strategies. They don't take that into account but the Boss all by itself discovered that trading rule and was pretty smart about it. Also, as far as the ranking goes, very interesting to me was not only was it looking for the highest momentum off of. It looks like about a six month lowest close, that was one of the highest weighted for the ranking. So basically, it's trying to find stocks that blasted up the most from their lowest close over the past about 6, 7 months. But it also was looking at a low score in about the same period, very interesting.
So, it's combining high momentum with some bad momentum as well. You know, not too much of this is something that I haven't discovered as a human trader and I’ve been doing this. Especially the ranking part I've been trading that way for the last seven years. And that's one of the biggest flaws I think in in other software platforms is that, they don't really allow for ranking. Most of the time, they're kind of using 1980s and 90s technology and really, you're optimizing your trading strategy on one symbol and not really looking at ranking. Which is a big mistake, because you're leaving tons of alpha on the table there.
But one of the biggest surprises that I keep on running into over and over, as I create trading strategies using the aid of the Boss is that. It'll go ahead and use stop losses I had never thought about before. So, once again index stop losses are coming into play in this particular strategy. What I told it as far as fitness goes is, and this is again referring back to what I talked about at the beginning that we don't want to have too much direction from the top down. Like we don't want to go ahead and add a ton of indicators into the buy and sell filters and ranking. We want the Boss to do a lot of the heavy lifting for us.
The only thing that I added in this particular strategy manually is I said, “Hey, the qualifier is I want at least the volume the dollar volume to be at least a million, over the past five days.” I think that that's pretty important. And then as far as top down where again the example I used before was, hey I want to go ahead and give tax incentives to the population and then they can just do whatever their own thing, right. They know what's best. And one of the stories that I really love that Milton Friedman didn't invent the story but told it. That was probably the first time I heard it was I pencil. And it was basically saying, like how the hell do you make a pencil.
There was just all these people that basically are doing their own thing and figuring out how to get all the pieces of a pencil together to sell it to you. There's just no way that you can do top-down economies like that. And people like Aldous Huxley that basically said, we need a technocracy and top-down approach to everything. Because science is so complicated, was completely wrong, wrong wrong. As many of these intellectual but idiots are. Anyway, getting back to the fitness function over here the top-down approach. The just the really broad strokes I gave it was I wanted to have a max drawdown goal.
Which is your you know let's say in 2007 to the bottom in march 2009. Warren Buffett lost 56% for Berkshire Hathaway. And so, I don't want to have 56%, I couldn't trade that there's no way. So, I set a goal of 25%, that's pretty reasonable. I can trade through that. I have a competent growth rate goal of 80%, which is pretty huge. Winning years 100 and winning months the goal was 80 winning months. And the strategy that it came up with was very interesting to me. It has a portfolio box score of 41, compounding on the growth rate in the simulation of 49 annual growth rate. And the max drawdown turned out to be about 30%.
I’m sure we could find a better strategy than this one even. Average position gain is 6%, which is fantastic that way commission and slippage aren't going to eat away at your profits. And that's a great strategy, absolutely fantastic. That was tested from 1997 to 2020 and the last 30% were out of sample. Again, if you join me on the website you can see this bad boy in action and it's got a beautiful out of sample and in sample equity curve with giant gains during the Corona crisis. You couldn't even, there's barely a re this is like a tiny little blip that was really fast and then rocketed right back up. Making way past equity highs of back when the Coronavirus hit.
So, that's exactly what I want to go ahead and see. And again, this was actually done live. I was on a group call and the first couple strategies that I ran, the Boss didn't really find anything. That's why I always suggest using a small population size and just run a few simultaneous tests. I think I ran like 4 and then I went ahead and pressed start one more time it was about 15 minutes from the close of our group meeting on ‘How to use the Boss’, and just helping people out. Of course, it designed this on the fly, so that was pretty cool.
If you're on the podcast listening to this right now, what I'm looking at is just a beautiful equity curve both in sample and out of sample. Exactly what you want to see, and what's really cool about it too is that for a lot of people, there's no one size fits all. I'm like you I've subscribed to 20 or 30 newsletters over the years. Typically, get burned by them because they don't have a real strategy it's not scientifically proven. By scientifically, I mean it has to be repeatable. You have to be able to send these rules to somebody on the other side of the world in whatever country right, it doesn't matter.
Anywhere, you have to be able to send them these exact rules and then they should come up with the exact same results, that's science. Now what's not science is drawing a bunch of lines on charts. It's way too subjective, what's the head and shoulders pattern. It's all garbage as far as I’m concerned. As soon as I realized that and I learned that hey, when it comes to trading you should look at stock trading like you would sports betting or playing cards like blackjack. Where you're always computing your odds of success. If you don't know your odds of success and failure and you're the size difference between your winners versus losers, maybe you shouldn't be trading. Because that's certainly not scientific if you do it any other way than that.
So, with the Boss, it does all the heavy lifting and of course tries all those different indicators and different settings to the indicators and just generates these things right on the fly. It never ceases to amaze me what actually works in the markets what generates a strategy like the one that I’m looking at right now, which is just a thing of beauty. And a strategy like this is very tradable. You know I was talking to one of our members Dan C out of North Carolina. He and his wife were talking about getting a motor home and retiring, because they've been using the trading strategies for many years.
They're very confident that they're going to keep working in the future here. I just feel they've just done so well since releasing back in 2014 is when I started this crazy game. It fits their lifestyle to trade once a month. But you know what doesn't fit their lifestyle, trading once a week, right. Because they're going to be retired, they don't want to have the stress of looking at the positions especially if they're traveling in a motor home across the country. You never know what kind of internet you're going to have. It might be a little bit sketchy. And so this seems to really fit their personality.
I would definitely not discount that. A lot of times when people when they stop following the strategies, because it just doesn't fit who they are as a person. Like for example, I think that the Rolls-Royce is an amazing vehicle and I think it is a beautiful vehicle. But it's not for me. I would never buy one, well I shouldn't say that. Maybe in my 60s it just seems to me to be an older man's car and I'm 45 so getting up there a little bit. So, maybe soon enough I'll have a Rolls-Royce but I just don't want one right now. The same thing with I've been I was looking at some boats and things and it just doesn't really fit my personality right now. Because I know the maintenance on those things.
I'm not worried about the cost but it's more of just having a pain, it's a pain in the butt, right. Everyone is different and so that's why if you can articulate like what I was talking about fitness functions. And you can articulate how much money you want to make, if you can approximate you know how much of a drawdown you can withstand. You know how often do you really need to win. You need to have a lot of winners your win percentage needs to be high. Because oftentimes you'll actually sacrifice. You'll sacrifice a lot of profits and leave a lot of profits. If you have to win let's say 90 of the time or 80 of the time.
That might not be so realistic, but if you are let's say a futures trader where you can have a winning strategy that only wins 30% of the time. But mentally, it could take its toll on you just keeping just staying with taking all those really small losses over and over and over. Then trying to let your winners run. So, there's a lot of different markets out there. That's why we have the Boss to go ahead and develop trading strategies on all these different markets all these different indices all these different stocks and sectors and so forth. It really gives you a broad overview of how to make money in the markets and trade a strategy that you can continue the trade month after month year after year.
With that, I will see you next time. Dan Murphy Signing out.
CLOSING: You've been listening to Artificial Intelligence, Real Profits with Dan Murphy. If you want to get the free report, Artificial Intelligence Real Profits, how to automatically build tailor-made Super AI Trading Strategies from scratch without writing a single line of code, go to ‘Portfolioboss.com/aireport.’ If you have questions or comments email Dan at ‘Portfolioboss.com’. We'll be back soon, until then this is Artificial Intelligence Real Profits, signing off.