Busy Weeks of Research

It feels good to get a blog post out after the past two weeks.  Imagine you’re at the airport in Detroit, Michigan, and your final destination is a small town in New Zealand.  You have multiple connections, several days of traveling, and the necessity of dealing with Covid regulations in every country that you enter along the way.  If one of your flights is delayed it could massively impact the rest of your itinerary, but it’s out of your control.  You basically have to submit yourself to the situation you’re in: eating whatever food is available, dealing with terrible sleep, lack of showering.  You just ignore how much it sucks, and focus on how happy you’ll be when you get to your destination.  That was basically my self-imposed past two weeks as I pushed through the uncomfortable and boring aspects of my latest machine learning for equities trading project, and I finally feel able to resume a good life again.

I think most of the year it’s possible to get a tremendous amount of work done, but still have a balanced life.  There are times for me though, especially near the end of a difficult project, where I have to push through to completion.  Working on anything creative and uncertain there’s the possibility/probability of giving up at all times.  Most of the time things don’t work how I want them to anyways, and I have to cast ideas aside.  I’ve been working on transitioning to trading stocks for months now, exploring everything from trying to trade once a day to once a minute, and I finally found two different algorithms that not only did really well on historical testing, but they also performed amazing during the past trading week as I tracked how they did.  The step from “it’s good on old data, and it looks good during the current day” to fully autonomous and automated was strenuous and harder than I thought though as usual.  But I did it, and had my first money making week trading stocks last week!  The dollar amount wasn’t large because I’m trading smaller quantities for all of February, but they made a lot of trades with a very large winning percentage in a flat market.   

I don’t want to jinx myself, but barring a major bear market or me doing something very stupid (both are possible), I might have enough with these two algorithms to set myself up financially by the end of the year.  I still have plenty of important tweaks and improvements to make over the coming months, but the massive data collection/preparation/research/automation part of the process is good enough now, and I can focus on how to make a little more and lose a bit less while my trading runs automatically.  I’ve learned so much recently about trading that has made the entire field so much more interesting and exciting too.  I could easily go full-immersion and think about trading 24/7 (I almost have withdrawals when the market is closed now), but I need to get back to a balance.

This week I’m going to do a bit of work each day (it’s admittedly nice that the market is closed today for Presidents’ Day), but I’m putting conscious attention towards health.  I had a good workout, breakfast, and meditation session already today, and that’s the plan this week.  Also reading more.  I’ve been glued to my screen obsessing over game theory/computer science topics, and it’s been hard to shift my brain into reading mode.  I have a great que of books, and I always feel better after reading.    

Hopefully my brain will refresh and I’ll have more lively things to share with you soon.  Hope you all are having a good start to your year as well!  I’ll share with you good or bad how things progress.

One Reply to “Busy Weeks of Research”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s