The Trend Is Your Friend, Momentum Investing - MaxSharpe

Momentum is the phenomenon that securities that have performed well relative to peers (winners) on average continue to outperform and securities that have performed relatively poorly (losers) tend to continue to underperform. The existence of momentum is a well-established empirical fact. The return premium is evident in 212 years (yes, this is not a typo, two hundred and twelve years of data from 1801 to 2021) of U.S. equity data. The empirical evidence is also available dating back to the Victorian age in UK equity data and in more than 20 years of out-of-sample data in 40 other countries in more than a dozen other asset classes.?

In this blog, I am going to share a systematic equity momentum investment strategy for the Indian markets with solid risk management followed at FinSharpe. I have also shared 16+ years of backtesting results free from survivorship and look-ahead biases. Such a long backtest gives us a prospect of performance during key crises including the financial crisis of 2007-08.

The Approach:

• Opportunity set/ Investment universe - I use the BSE 200 as the investment universe for this strategy. To backtest any strategy without survivorship bias it is necessary to have periodic constituents details for that investment universe from the start to the end of the backtesting period. We need this information to define our investment universe at that point in time. For example, during the backtest when the algorithm is creating the portfolio in January 2007 then it should take constituents of BSE 200 at that time else the backtest results will be inaccurate and inflated. For most of the NSE indexes, we don't have the periodic constituent's information for such a long period hence we use BSE 200.
• We use a minimum of one-year pricing data as a lookback to calculate momentum score and rank our universe. Such a long look-back period of one year is to ensure that we remove the noise and identify a stock that is structurally in momentum due to strong fundamentals and performance. For this purpose, we use factors such as multi-period regressions, return quantiles, and Hurst exponents of price time series. The strategy is rebalanced once in two months and one year look-back period will be on a rolling basis.
• During the backtesting at every rebalance date we need to perform the following steps: 1. Identify the universe at that time using periodic BSE 200 constituents information. 2. Filter out any stock that has less than one year of data at that time. 3. Use the factors I mentioned above and calculate a momentum score for each stock in the universe. 4. Rank the universe using the score calculated in the previous step in a way that the highest momentum stocks come at the top.
• Now let's take 50-60 stocks from the ranking and create a diversified portfolio of around 25-30 stocks.
• To create a portfolio of around 25-30 stocks we use both constraint?Mean-Variance Optimization?and?Hierarchical Risk Parity?in a way that the maximum weight for any individual stock is not more than 6-7% of the portfolio and sector weight is capped at 25% of the portfolio. We also use a commission/ trading charge of 0.25%.
• Now let's backtest the strategy using?Backtrader from 3 Jan 2006 - 13 Apr 2022.

Backtesting Results:

Here are our key performance metrics from backtesting the strategy against the Nifty 200. I have used Nifty 200 as couldn't able to get S&P BSE-200 pricing data beyond 2011.

It is a very impressive performance on key parameters such as CAGR, Sharpe, and Sortino ratio the strategy beats the benchmark by a wide margin. However, the Max Drawdown is quite high as compared to the benchmark. This was during the year 2008 financial crisis.

The above monthly return heatmap also highlights the fact that volatility during the financial crisis was significantly higher than the current time including during the pandemic.

The above table of EOY returns vs benchmark has only one year when strategy could not be able to beat the benchmark i.e. 2008. Overall the strategy has outperformed the benchmark by a huge margin and I can share many more performance parameters however don't want to make the blog lengthy. You are most welcome to contact me if you find this interesting.

"With a good perspective on history, we can have a better understanding of the past and present, and thus a clear vision of the future."?— Carlos Slim Helu

Please Note: This analysis is only for educational purposes and the author is not liable for any of your investment decisions.

Thank you!

References:

1. Fact, Fiction and Momentum Investing - Cliff Asness, Andrea Frazzini, Tobias J. Moskowitz

Know more about such strategies:

https://finsharpe.com/

1. Optimal Alpha -?Click here

2. Multi-Asset Portfolio -?Click here

3. High risk, high return -?Click here

4. Value Vision -??Click here

5. BlueChip Focus 15 -?Click here