I just came across an excellent and highly relevant piece of research "A comparison of machine learning model validation schemes for non-stationary time series data" by Matthias Schnaubelt. Features like non-stationarity, concept drift, and structural breaks present serious modelling

Few years ago I published two post trying to give simple explanations and intuition behind complicated formulas used for calculating vol indexes. However few of you emailed that some charts are missing from these older posts, and for technical reasons since I could not restore them, I decided to

I think if you're reading this blog, you're probably already a knowledgeable options trader, and have heard of maximum pain theory - an idea that market moves in a path that hurts ( causes losses ) most amount of market participants. In options it is typically stated that simulating

As I wrote before, Bitcoin volatility is quite different from volatility of other assets. I will continue with the same topic here. Bitcoin prices shot up to all-time highs this winter, and have sharply declined since. When an equity index declines, volatility typically moves up, but in the case of

If you're working with daily data (without access to intraday data) and need to calculate volatility, then using close-to-close squared returns is by far not the best way to go. Trades and quants know that it is a very noisy metric, and come up with few work-arounds. In this post I will do a

Reader Eli asked in the comments "I used to program in Matlab a lot in the past. Now I feel that Python is a better way to go. Any thoughts? Yes, quite a few - however I also hope to hear some ideas from other readers. Matlab is an excellent tool for analytics, and I have been using it for over

MATLAB Computational Finance Conference was a great event that I attended last year. There is a lot to learn for average Matlab user, and anyone working in finance. The agenda looks especially exciting this year. Ping me if you want to meet during the event.

Comment from reader Rich on yesterday's post prompted me to run a quick investigation - does adjusting bars helps to bring distribution of highs and lows closer to theory? The answer is definitely yes, but as everything else in life the reality is complicated. Here's what I found: Using

Just few days ago stock.nu published a post with charts showing that empirical distribution of daily lows follows a U-shaped pattern, i.e. daily low is not equally likely to happen at any time during the trading day, rather low is more likely to occur near the open or the close. Similar (symmetric)

The Data Science Institute at Columbia University and Bloomberg are holding a workshop on Machine Learning in Finance. The presentations look interesting and the price is right - just $30 if you have a valid student ID, or $100 if you don't. Some research articles are already available for

Notes From Academia: Exercise Boundary Violations in American-Style Options Here is the link to the research paper - Exercise Boundary Violations in American-Style Options: The Rule, not the Exception Abstract: An exercise boundary violation (EBV) occurs when the current bid price for an American

Different commodity including some volatility and bitcoin futures settle to an average value. For example VSTOXX settles to 1 minute sampled average value of the index in the last half hour of trading, Bitmex XBT/USD futures settle to 1 minute sampled average value of the last two hours of trading,