Category: Collections

Curated list of posts on a specific topic.


Some portfolios you consistently ended up with higher returns, implying that there was something about the market, something systematic, that was driving them. What are these factors? Factors, Intro

Systematically accounting for excess returns became an academic sack race. Factors, The Famous 5

While fundamental factors play a role in explaining excess returns, technical factors cannot be ignored either. Thus, Momentum. While the market can be sliced-and-diced in many different ways, here’s a simple way to go about it: The All Star Backtest

Momentum portfolios are extremely volatile. But, is it possible that a portfolio of less volatile stocks out-perform the market? Ergo, the Low Volatility Anomaly.

Both Factor Rotation (buying what worked best in the past) and Multi-Factor (buying all factors) work. As long as you stick with it.

Fat Tails


Years of returns can get wiped out in a month in the markets. While investors mostly focus on the average, the tails end up dictating their actual returns. (Introduction)

Sampling and Measurement

Typically, a uniform sample is taken. The problem with this is it under-represents the tails. This leads to models that work on average but blow up on occasion. One way to overcome this problem is through stratified sampling. (Sampling)

Expected shortfall (ES) is a risk measure that can be used to estimate the loss during tail-events. (Measuring)


All assets have fat tails. It is a feature, not a bug. (Historical)

SMA Over Indices

Simple Moving Average (SMA) is one of the oldest and simplest measurements of trend. Arrived at by taking the average of prices over a period of time, it remains a popular tool for timing investments and risk-management. The following series of posts outlines how investors can use SMAs to get superior risk-adjusted returns.

SMA Strategies using ETFs

SMA strategies that use ETFs to create trend-following portfolios.

Reducing Drawdowns in SMA strategies

Shallower drawdowns allow a bit of leverage to be employed. This could be a good starting point for a NIFTY futures trading strategy.

Slopes vs. Cross-overs

A lagged response will result in higher drawdowns. It could, however, lead to lower transaction costs by papering over short-term mean-reverting moves.

Transaction Costs

Transaction cost analysis to backtests give investors an idea of what gross and net returns of different SMA look-backs look like over buy and hold.

Long-term Returns

Strategy outcomes depend on the underlying index and holding-periods. There is, alas, no magic formula.

Global Equities Momentum

A brief introduction to Global Equities Momentum and a look at various alternative scenarios.
Read: Part I

Could it work on value indices?
Read: Part II

Swapping momentum at the final step boosts returns significantly.
Read: Part III

Averaging out returns over different formation periods boosts returns and reduces drawdowns.
Read: Part IV

Track the virtual portfolios we setup using these strategies and follow the trades on our Slack channel.
Details: Trades and Portfolio


Can global “macro” data be used to model local markets?

Correlation and linear models

A series of posts exploring the (lack of) linear relationship between USDINR, OIL and the NIFTY 50. Part I, Part II and Part III.

Machine learning

Using an SVM over dollar indices to trade the NIFTY

An SVM can be tuned with different parameters. Principal among those is the kernel to be used. Part I looks at the difference in predictive power when different kernels are used to train an SVM over the same dataset.

Part II tries to further tune the most promising kernel from Part I – the polynomial kernel. However, we find out that there is no silver bullet.

So, we take all the dollar indices (including USDINR) and observe the predictive power of an SVM using a polynomial kernel with varying degree parameters. We find that only two out of four indices are actually useful and that going too far back in history while training the SVM is counterproductive (Part III.)

Part IV trains an SVM using the two promising indices and their respective best-performing degree parameters from Part III. An ensemble model that chains the predictions looks promising.

While performing the experiments above, we noticed that none of these models side-step the 2018 drawdown. Their principal limitation is that they are “macro.” They will not handle local events well. So we combine the model from Part III with a Simple Moving Average and find that it leads to lower drawdowns in long-short portfolios. Part V also leads us to conclude that using just one of the indices is sufficient to meet our twin goals of shallower drawdowns and higher long-short returns.