Back to Work

Another solid week here in PR.  I was all excited to blog about it yesterday (Sunday) morning, but I went paddleboarding, had a great coffee, started reading, and then it was evening.  I’m currently in the  middle of my normal “workday”, but getting this out felt more important than my immediate research.

When reflecting on my week there were no extraordinary accomplishments or events, but it was one of those incremental improvement weeks that make up the majority of a good year.  I pushed hard on my three workouts (legs, upper body, and one sprinting day), I ate very healthy most of the days (a few weekend treats, the first of which felt mentally needed), and I didn’t go crazy on my caffeine (but had a second serving a few days). 

I lost a pound, and noticed that paddleboarding was much easier, as well as my single-leg squats today.  I think my balance and stabilizer muscles are improving from my training.  My body Friday morning was much better than Sunday morning (from the ice cream unfortunately), and I think I’m ready to more frequently cut out weekend junk so I don’t have to work off the bloat in the beginning of the week.  

In regards to trading I got my stock trading running again!  I found a small technical issue with my machine learning that doesn’t make sense to me, but was easily fixable.  I was wondering why my research was showing such consistent gains historically on my strategy, but I didn’t make money when I tried it for a few days.  There was an element of bad luck/small sample size, but I also found when I saved my algorithm using all of my data to train it on the final version, it wasn’t doing as well as when I held some of the data back.  I noticed that different variations of the machine learning were predicting opposite from each other (one would predict up and the other down on a stock price) which was odd, but once I made that tweak they are now almost always in agreement (and the results seem good too!).  

On the negative side the equities data provider that I paid for and have been using for the past month for research seems to have major issues.  Missing data, and occasionally wrong data.  I trusted that the data I was receiving was good, and it’s only occasionally wrong, which is more scary because it’s hard to know when it’s wrong.  I read about other people’s problems, then looked on their Github and saw how many issues they have.  I was looking to get some CME futures data from a different provider when I went down a rabbithole when multiple people mentioned their problems with my stock data provider.  Who knew it would be so hard to record accurate stock data!?  Apparently the majority of financial data providers have issues with their data not consistently being correct or accessible.  I found an old school company that has been reliably working for years, and is priced very reasonably, but I’ve avoided it because they have a clunky website/you have to use Windows to get their data (I’ll have to download some programs to make it work on my Mac and Linux computers).  My goal this week is to onboard myself with their API/data, get comfortable manipulating the data I pull from the site, and recreate my equities data sets (which I’ve annoyingly had to do far too many times).  For quantitative trading this is too important.  As much as I want to push forward developing more algorithms with my “good enough” datasets from my previous provider, I know taking a step back and improving my data is necessary first.  I can still trade once a day, and I’m learning a lot from that.

Bitcoin was/is super wild this past week, but I don’t like to think about it.  I have a very refined algorithm that runs and makes decisions without any input from me, so for me to watch the prices go up and down is a total energy-suck/waste of time.  The week before last my algorithm lost a bit of money, but then made it back this past week.  This is the highest volatility since Covid, and it’s a great time for trading assuming risk management is dialed in, which I feel good about.  I realized I’ve been spending too much time watching the prices like a slot machine, and I need to ignore Twitter and other sources of media that are constantly in manic states of euphoria or panic.  

I’ve complained about some things living here in Puerto Rico before, but these past few weeks have been amazing.  Perfect weather, Covid restrictions have been lifted enough that everyone can have fun again, and in general it’s such a great place to live a balanced life.  I can swim, workout outdoors, go to a cafe, work a good amount, greet the various people from the neighborhood on my daily walk, and then have a relaxing night. 

Back to the research and end of my trading day!  Leaving you with a pic of me, my gf, and our friend, and another guy here that was having a hard time cutting the cheat meals this week.

One Reply to “Back to Work”

Leave a Reply

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

You are commenting using your 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

%d bloggers like this: