I am researching trading strategies, particularly involving time series analysis, but also machine learning, on data sets that range as far back as 1950 and would like to compare the accuracy of various models, up to ~25 at a time.
My for loops in R for an ARIMA + GARCH model of the S&P500 values starting from 1950 to today take about 10 hours to evaluate all models with p and q both ranging from 0 to 5.
My specs are an AMD Intel i5-8600K, 16GB of DDR4 RAM, and an AMD R9 Fury, but I am not using any parallel processing.
Thank you.
It depends on the type of analysis. I do it on i7 with 16 GB RAM and it works fine.
Jordan Hubbard ,
This is a tough question. There are many possible solutions.
First, without using parallel processing, there's not much you can do to speed up your code in terms of personal hardware (considering the code is already optimized for single threading).
Personal computing processors nowadays are physically limited to ~4GHz.
Your gig is already pretty good, and you'll not gain much more by using a faster processor.
Since you are testing multiple datasets, I suggest you use parallel processing to make use of all your available threads. This will be the best way to gain speed in your tests. I always try to do this.
Also, I don't know if there are many libraries in R with GPU array operations for ARMA and machine learning models (excluding deep learning), but you could try to use those (if your GPU is supported by such libraries).
Finally, R is usually not the fastest programming language. Julia has gained attention lately due to its high processing speed.
- Badges
- Science topic