First of all, Algominr is a great tool. I have some basic coding skills, but it probably will speed up testing and prototyping of strategies for me.
Importing other data than prices will make the tool even more useful, but at the moment, the handling of price data itself is quite difficult. Maybe I am missing something, but from MT4/5 brokers I can fetch one year of 1M data at best (the ticks in MT5 are less restricted, but they don’t seem to be imported by Algominr).
One year is way too short to test a strategy. I also tried to create a custom symbol in MT5 and import my own data (into MT5 and from there into Algominr). Apparently, custom symbols cripple the symbol list in Algominr (it is twice as long as it should be, with a second half of empty records). The import fails (“instrument chart should be opened at least once”), even if I had browsed the 1M data in a MT5 chart. A custom price/tick data import (like from csv) would be great.
I am not sure, whether it is an advantage to link the data to algos/projects. I was surprised to see that and I like it, because it keeps things separate. However, with larger data sets it might become inconvenient.
first of all, sorry for the late reply, I’ve been very busy over the weekend.
Great you like Algominr!
Are you talking about the data that you get in MT4? Because yes, MT4 is restricted in what it can access for historical prices. It’s better to go through a brokers/data providers api. I will add a lot there in the future. And I will also add non-price data to Algominr.
.csv import is also planned.
Thanks about the tip regarding custom symbols. I will check that out.
Linking strategies to data does usually make sense, as the strategies that work for one symbol do not necessarily work for another. Additionally, this way you can use for example several inputs to predict a price (e.g. predict the EURUSD based on gold prices, AUDUSD, US100, FTSE100, etc.)
The project is still very much in development and many features will come this summer.