Algotrading journal

Peter Pauer

Peter Pauer

Hello,
this si my journal of algorithmic systems portfolio.

My aim is to share transparent (real) results and find like-minded traders in the process to exchange information and tips.

I will post first some information I want to share. These are usually the questions I get from people when I talk about trading. So I will answer these beforehand.

My background
I am still working outside the trading industry in C-level management. Sadly I have no programming background. I did study economics and I have master's degree from marketing. I live in Slovakia, Bratislava.

I am trading since 2009, but with the current algo approach since 2018.

I am not a native speaker so plesea :) forgive my grammar.

Hardware
I am using VPS to run all of my scripts and connect ot my broker's trading platform. I am using my PC from home to conduct all analysis and research. The PC is nothing special, I build it myself from mid-to-high level components. It is AMD Ryzen 5 3600 6-Core with 16GB RAM, I am using M.2 NVMe SSD disks. I am writing this because some people are convinced that you have to rent at least a super computer in China to do some backtesting.

Software
Amibroker - I am using Amibroker for a) backtesting, b) market scanning
Python - the whole automated system is coded in python. I am using it to scan the market, send entry and exit signals to my broker's platform via API and to save the outcome of the trades in a trading journal.
TWS - I am using Interactive Brokers as my broker. I am using TWS a) to send orders via API, b) to visually check on my positions

There are plethora of other applications and software I am using, which help me but are not essential for my trading (Windows as OS, OneNote as note taking software etc.).

My approach to trading
I am trading a portfolio of algorithmic (just a fancy word for automated) strategies which trade US stocks. My goal is to build a diversified portfolio of non-correlated algo strategies (...and lay on the beach and sip piña colada :) ). I am trying to diversify my portfolio by using different trading strategy approach - breakout, mean-reversion and seasonal. I am using also different timeframes - from intraday until holding positions for couple of months.

I would like to add more diversification in the near future by adding more non-US stocks and other instruments (mostly futures and options).
 
My trading results
Fundseeder
I will be sharing my results by using two different formats. In the first one, I will be sharing screenshot from fundseeder.com website. This web was co-created by Jack Schwager and it gives traders credible option to show their track record. It is connected to trader's brokerage account via API and therefore it tracks every change in the trading account.

I connected my account to the fundseeder on January 1st 2021. My account is nominated in €. I am currently trading 6 figures (not big, not tiny).

Results as of 31st May 2022 (last update)

FS_31_5_22.PNG


Google sheets
The biggest disadvantage of FS is that all the statistics are based on the NAV (Net Asset Value). In other words, what is the current value of all the positions you are currently holding in your account. This is a great metric, but it if you want to measure the pure performance of your strategy it is not the best option. Firstly, because all the costs are deducted from your account and therefore they are calculated into NAV - market data fees, broker interest fees, etc. Secondly, because the NAV metric logically calculates also the assets which you are holding long term and are currently in the red numbers.

I would like to measure also the performance of each of my strategies (something I cannot do with FS) and I also want to see the performance of closed trades (Cumulative P&L). Therefore I am also using python to extract all finalized trades and I am using google sheets to analyze them.

The text above should answer your question: Why are the results in FS different than the ones you are showing in your other pictures.

Here is the latest P&L result as of 1st of June 2022:

GS_portf_1_6_22.PNG
 
@felix_arb
Hi, thanx.

I am trading portfolio of strategies. Each strategy has intentionally different settings and uses different timeframe.

- LWI is strategy based on seasonality. Avg. days held 77.08 (backtested on data from 2000 - 2017)
- MBA is a breaktrough strategy. Here I am trying to enter into a trend in it's early stage. As a result I am losing money on most trades, but when I hit the right one I usually ride it for a big profit. Avg. days held 36 (backtest 2000 - 2017).
- CLX is another breaktrough strategy, but this one is littlebit faster. Avg. days held 4.22 (backtest 1990 - 2019)
- PMS and PML are mean-reversion strategies. Same strategy - first is entering in short direction, the second in long. Both are quite fast. Avg. days held 2.76 (bcktst 2000 - 2015)

I am currently working on intraday strategy (it will not hold positions overnight). I am running it so far on a demo account to test everything. Hopefully I will launch it live in couple of weeks. I tried to post a video here, but unsuccessfuly, so here is a pic at least:

rt_scanner.PNG
 
Last edited:
You must log in or register to reply here.
Back
Top