Search results

Because it's not really API you are installing, it's the gateway or TWS that the api connects to (unless you are using FIX). IB architecture is reallly frustrating but that's just how it works.

Hello Ben, If you are a programmer and not interested to build your own backtester (which I have done many times), it's worth looking at quantconnect, you can get decent data (I think they even have tick data) and runtime environment for a small price to easily test your hypotheses with good...

I'm actually using quite sophisticated neural network ensemble to predict "optimal" long/short portfolio, this is just the entry method I'm currently using (getting open prices daily + generating parameters based on new data combined with historical data + actual 30 separate neural network...

These are daily trades, each set of trades is triggered at approx ~5 min after market opening. Depending on the predictions of the system the holds positions are held about 1-5 days, mostly the shorter side (1-2 days)

Problem with stop limit is that it doesn't trigger if price goes to "right" direction. What I have been thinking is conditional peg to midtpoint that is triggered by price action. However I'm not sure if this will get me better or worse execution.. I also know that this is almost impossible to...

Hello All, I have a working system that I use to trade faily liquid (>5M volume/day) NYSE/NASDAQ stocks. I use opening range breakout to enter the trades based on my system (using interactive brokers API). I dont't trade that big, my typical position is 1-10k usd so nothing that moves the...

Thank you Mark, it then looks like the data is indeed actual market data, not missing data.

Thank you very much if this is true! Could someone verify if their data has the similar prices at 2020-03-18 or at monday 2020-03-16 -tuesday 2020-03-17 night

Okay I ordered the dataset. Lots of the data seems to be okay (I don't have another high quality dataset so I can't cross-validate) but there are points where there is missing and/or invalid data. Some examples attached. (The charts may be a bit messy as they contain some of my proprietary stuff...

Algoseek is about $550 per 10 years of data for single symbol while firstratedata is 200 euros for 70 futures contracts since 2006. There is no survivorship bias in futures The price point was the main reason I want to know specifically about firstratedata (unless you specfically know of other...

Hello, I'm looking for 1 minute (second resolution would be even better) data set for futures strategy backtesting. Firstrate data seems like an obvious choice from the price perspective. I didn't find any information about their data quality anywhere, has anyone done any comparisons to...

I want to compare my results to "random" results to test null hypothesis. Because random methods rarely work at all because of commissions etc, what is the optimal way to construct the random dataset that I compare results to?

That's true. I'm just too deep into my usual algo "get all the cents every trade" thinking that I can't think this straight. Thanks for your comment.

Thank you, this is about the only sensible answer (with the option comment) There is always a premium and a larger spread in longer term contracts (and lower liquidity), I'm trying to find optimal risk/value contract - in your opinion a longer term contract would be the optimal instrument for...

I'm thinking of doing a long term (multi year) speculative trade into crude oil and natural gas. What would be the easiest/most effective way to do this? With futures there are rollover costs etc and with etf's there are other costs and all kinds of risks involving bankcrupcies, regulations...

I would also like to hear yes/no answer to this question ASAP. It's a kind of a red flat to me when representative says "I'm in hurry so I can't aswer any further questions now" when a very serious issue is raised by a press and then wanted to be clarified by a huge amount of their customers here

You can try different sampling methods, ie. instead of using totally random sampling, add some clustering variable to the mix. I have personally used a method where instead of using random points from the curve, I just use the next datapoint for random X% of points.

Hello, I'm trying to test a portfolio of strategies with longer term data and I only have 1 minute data. All my orders are limits (with some stops as security measures). What would be a reasonable way to model fills? For example for buys, I currently model a fill if the 1 minute bar...

What does "density" represent at the curve?

On trading simulations using monte carlo usually refers to randomizing your trade sequence to see the possible risk/reward combinations because there is "not enough" data available, I'm not sure if this is what talon means but it's not about randomizing market data, more about randomizing your...