The Hard Lessons I Learned Using Machine Learning to Predict the Markets
Years ago I started Neural Market Trends to use Machine Learning to predict the markets. I wanted to use all these fancy algorithms to predict the direction and outcome of the stock market 1 week, 2 weeks, and 1 month into the future. I would execute trades on what it said and then get rich. Well, that didn’t happen. Instead, I lost money.
I ran this machine learning experiment from 2007 to about 2008, just as the market started to blow up. I remember meeting with a professional options trader about a model he wanted to me build for him and he warned me that something big was moving in the market. He didn’t know what it was but something was happening below the surface.
That “under the surface” movement turned out to be Bear Stearns imploding. That threw the market into a complete tailspin and things have never been the same since. The market shenanigans in today’s market are a result of the massive bailouts and QE liquidity that the Federal Reserve and all US financial institutions created.
The sad thing was that they had to do that. Things were so bad and we were inches away from total financial collapse. What would’ve followed would’ve been a global depression. It would’ve been one helluva hangover. Luckily it didn’t happen, we just kicked the can down the road for a few more decades.
I changed my investing and trading philosophy after 2010. I decided to stop day trading stocks and Forex. I did ok day and swing trading, but the saying “trading is the hardest way to make an easy buck” rang true. By the time you subtracted the trading fees, slippage, and taxes, you didn’t make much money for such a small account I had.
To quote Macbeth, it was “…a tale, told by an idiot, full of sound and fury, signifying nothing.” I expended a lot of effort to make a few bucks. It wasn’t worth it in the long run.
So I dropped my machine learning models in favor of long-term investing. I picked up A Random Walk Down Wall Street and it changed how I invested and saved money.
Once I did that, my wealth exploded, but in a good way.
I spend my time dollar cost averaging into a 401k, diversifying, rebalancing, and reinvesting all dividends and capital gains. I don’t pay attention to the market that much and I like that pile of money gathering momentum till it reaches critical mass.
Once it hits critical mass I’m outta here, I’m retiring and going hiking.
Over the years I tinkered with Python code and rebuilt someone else’s code I found on the Internet for a Forex Trading Bot in Python. It was a way for me to teach myself Python and I wanted to get better at it.
After a few hours, I got the code working and connected it to a paper trading account at Oanda.com. Then I let it run.
I can’t tell you the joy I felt when I watched that Bot execute buy and sell trades. It executed stop losses and closed trades at the end of the trading day. Good thing this was done in a paper trading account because the damn thing lost money.
So I built a few different technical trading strategies, installed them, and let the Bot run again. They worked but lost money.
The entire Python Forex Trading Bot was a proof of concept that worked but lost money.
I played around with this because of one single reason, the markets sing a song that is hard to ignore. Whenever there’s volatility and the market is crashing or shooting higher, I get excited. I would make money off volatility and love the chaos.
Chaos is a ladder, at least in the financial markets…
For example, I took part in the original GME melt-up and made more money in two days than I did in 6 months of working. This time, it was easy money, but I knew better. It wouldn’t last.
The GME frenzy would trap a lot of people and it did, my partner being one of them. By the time she caught wind of the excitement I was already out of my position. I tried to discourage her from buying but she didn’t listen.
She went long and got trapped. It wasn’t any crazy amount of money, just a bit of mad money she had in her IRA but still, it stung.
I get that. There were times when I blew up my account regularly because I wasn’t learning the lessons that the Market was trying to teach me.
In German, we have a saying “wer nicht hört muss fühlen.” That translates to “if you don’t listen then you’ll feel it.”
The Market lures us in with her siren call of easy riches, Lambos, and an easy life. Then, when she dashes you on the rocks with your meme stocks or shitcoin, you’ll have to trudge off back to work feeling like a degenerate gambler.
That happened to my friend. He took out $100,000 on a second mortgage to day trade stocks, only to lose it all in a bad market selloff.
Been there, done that. I learned my lessons.
Over the years I realized that the markets are nothing but a giant money simulator. Fiscal and monetary policymakers, politicians, and countries monkey around with the settings of this simulator and watch it go up and down in value.
Thousands of talking heads, myself included, would write and talk about how the market’s momentum is increasing or decreasing. We’d tried to figure out if the price of gold would go up with the latest rise in interest rates.
Eventually, some lucky guy will write a book about how they made their millions doing some strategy. They’ll make money off their book but never realized survivorship bias made them rich.
Yes, survivorship bias is real. It happens.
It doesn’t happen much if you’re a short-term trader but it happens a lot if you’re a long-term investor. The long-term investor wins because the magic component of wealth building is time.
Rather, it’s time and dollar cost averaging into a diversified portfolio of low-cost assets like mutual funds, index funds, or ETFs.
Sure you can have some individual stocks, but I leave that for any “mad money” I have and for when I hear the siren song of the Market singing of an imminent melt-up.
My friend who lost the $100k quit day trading and recouped his losses over time. He started a successful business and instead of day trading, he built a business that generates massive income for him and his family. Does he invest in a retirement account? He sure does. He makes more money over time than he did before flipping hot stocks. That’s winning in my book.
My partner? She’s still long in GME but makes more money from dividends in her 401K than what she lost on paper per quarter. She adds the maximum to her 401K every year.
Me? I do the same as my partner but I stay away from volatility now. Yet, the Market call is very hard to ignore. She sings of easy money and it’s tempting. My old age tempered with half-assed wisdom keeps me away, for now.
Time will tell if I dip my toe into the market again.