Recent content by kojinakata

  1. K

    Next Steps after Building Strategy Pipeline - Advice for Getting Hired

    I am not from the US, but went to a "top-tier" university where I am from, top 5 to be exact. The why is that I always loved statistics, machine learning, finance and trading. I want to do what I love. At least, I want to try to get into a job I love.
  2. K

    Next Steps after Building Strategy Pipeline - Advice for Getting Hired

    Hi elitetrader, I have coded a reliable and sufficiently realistic vectorized backtesting script. It is mostly based on quantstrat and related packages in R. I have tried to implement methods to avoid overfitting based on techniques written in papers that most of you might know. I am testing...
  3. K

    Dukascopy Historical Volume Data Spikes

    Thanks for the link, seems Finam is a good source with volume data. I will have to crosscheck the volume with another source though.
  4. K

    Dukascopy Historical Volume Data Spikes

    The thing is Dukascopy is free and I don't have the resources to buy data from another source. However, volume charts from Oanda doesn't seem to be similar to the Dukascopy one and I can't seem to download Oanda's daily data.
  5. K

    Dukascopy Historical Volume Data Spikes

    Hi everyone, Just downloaded historical data from dukascopy, but when I plotted Volume against time, I think there are a few irregularities and problems in the data. I donwloaded 5 minute bar data of BID prices from 01.01.2013 to 31.03.2017 for EURUSD pair. I have attached the plot of Volume...
  6. K

    Parameterless Strategies

    Thanks for the answer. I am not sure I am able to understand your reply clearly but maybe you can tell whether the following strategy can be made a parameterless one: There are 3 signals: 1 - MACD cross, 2 - SMA cross and 3 - EMA cross. How do I use weightings to minimise correlation between...
  7. K

    Parameterless Strategies

    Recently I have come across something that I couldn't grasp. In this chat (https://chatwithtraders.com/ep-103-dave-bergstrom/) the speaker talks about parameterless signals to avoid data mining bias and overfitting. However, I am having trouble grasping the concept of parameterless rules... Even...
  8. K

    Systematic/Quantitative Trading Testing and Analysis Environment

    As a side note: I emailed Mr. Thomas Wiecki about the blog post on Quantopian (http://blog.quantopian.com/parameter-optimization/), and I am waiting for an answer. The blog post was closed for comments.
  9. K

    Systematic/Quantitative Trading Testing and Analysis Environment

    My most basic trade is as follows: I use MACD crossover as an entry to short/long position with a %2 risk calculated by the distance of the stop loss and the position size. The exit is just closing the position either by a trailing stop or a reverse signal from MACD. I don't reverse my position...
  10. K

    Systematic/Quantitative Trading Testing and Analysis Environment

    Thank you for your answer. Seems that retail platforms are inferior to what is capable with one programming his own platform. I just could not grasp what you were asking in the first paragraph? I define a trade as entering and exiting a position (round-trip). I define transactions as entering or...
  11. K

    Systematic/Quantitative Trading Testing and Analysis Environment

    You were right that it is capable of Walk-Forward analysis however, the blog post doesn't mention any capabilities towards generating 2D/3D graphs to see whether the optimized value has persistence in nearby values (whether the optimization returned a "peak" or "plateau" value). I am going to...
  12. K

    Systematic/Quantitative Trading Testing and Analysis Environment

    I don't think Quantopian can do Walk-Forward Analysis or generate 2D/3D graphs to see whether the optimized value has persistence in nearby values (whether the optimization returned a "peak" or "plateau" value) or do Monte Carlo Simulation.
Back
Top