Category: Investing Insight

Investing insight to make you a better investor.

Correlation vs. Returns & Volatility

0

The previous post presented historical NIFTY 50 constituent pair-wise correlation distributions and discussed how high correlation environments persist. How do the 5th-quintile return and volatility look like?

If you focus on the 5, 10 and 20-day pair-wise correlations (T5, T10 & T20), you’ll see large left-tailed returns and high volatility in the 5th quintile compared to the others (1st, 2nd, 3rd & 4th).

Looking at this the other way, T10 – the 10-day pairwise correlation – has this profile:

It looks like if you dodge the 5th quintile here, you might be able to boost returns in linear strategies.

Categories Tags ,

Index pair-wise Correlation

0

Sometimes, it feels like all stocks in an index are moving in the same direction. Very rarely is there full “chaos.” Sometimes, even when there is some dispersion, it is overshadowed by larger moves in a few index heavy-weights. To get some intuition around this, we calculated the pair-wise correlation of the NIFTY 50 constituents since 2015 across different lookbacks, bucketed the median daily correlations into quintiles and plotted them.

If you do an rle, you’ll get an idea about the number of consecutive days spent in each quintile. It looks like quintile 5, representing a highly correlated state, is sort of sticky.

For example, if you zoom into the 20-day lookback distribution, notice how quintile 5 stands out.

It appears that low-correlation environments are actually not that sticky. So, if you see a quintile 5 form, bet on it lasting a few days.

Categories Tags ,

Hedging Momos

0

Previously, we discussed how momentum itself trends and how that can be used to manage risk. Using simple moving averages showed promise when it came to some versions of our slow momentum models (see Trending Momentum Models). However, given the faster turnover of our Momos, it wasn’t a suitable approach (see Trending Momo Models).

What if we shorted the NIFTY to hedge against market-risk instead?

The naïve approach, tagged “HEDGE_FULL” above, shorts the NIFTY in proportion to the rolling beta of the strategy. Turns out, this is a very sub-optimal way to go about it. Hence “HEDGE_SMART”, that tries to minimize the basis risk inherent in this approach, adds about 3-4% to the strategy’s returns (likely eaten away by transaction costs & taxes) and reduces the max-drawdown by a significant amount.

The question is whether the benefit of lower drawdowns is worth the added cost and complexity? In the case of Velocity, it could be.

Related:

Categories Tags ,

Momentum Portfolio Size

0

Previously, we looked at skip-months, rebalance frequencies and formation periods for momentum portfolios. A 1-month skip & monthly rebalance turned out to be ideal. However, the most popular 12-month formation period is “magic” – not a terrible choice but not super scientific either. The only thing left to toggle is the portfolio size.

A 20-stock momentum portfolio seems to be the ideal config.

This is pretty much the standard direct-equity momentum portfolio: 12-month formation, 1-month skip, 20-stocks with a monthly rebalance.

Code and chart on github.

Categories Tags ,

Momentum Formation Period

0

Previously, we looked at skip-months and rebalance frequencies for momentum portfolios. A 1-month skip & monthly rebalance turned out to be ideal. However, we did these analyses keeping the formation period the same at 12-months. What if we changed that as well?

Turns out, there is no single “ideal” formation period where all stats converge. However, if set the rebalance frequency to 1-month, the average of the formation periods of the top performing portfolios works out to 12-months.

If the momentum fund is large enough, then it could probably be sliced into multiple sub-portfolios, each with different configs to avoid this magic 12-month formation.

Also, since the underlying process creating these portfolios is the same, the equity curves come out all bunched together. There maybe differences in month-over-month performances but they are all highly correlated.

Code, charts and statistics on github.

Categories Tags ,