Category: Investing Insight

Investing insight to make you a better investor.

Is there a correlation between USDINR and Tech stocks?

shyam May 2, 2015 0

tl;dr

No.

The Myth

Regular viewers of CNBC might have heard the refrain that “IT stocks went up because the rupee went down.” But it turns out that it is the classic case of the journalist/reporter/anchor going in search of a reason to explain a random market event. If there is such a correlation, then a scatter plot of USDINR returns vs. CNX IT index returns should unearth it.

Scatter Plots of Returns

Weekly:
CNX IT.USDINR.scatter.weekly

Daily:
CNX IT.USDINR.scatter.daily

As you can see, there is no obvious link between USDINR and technology stocks. But what if the effect manifests after a lag?

Cross-Correlation Plots of Returns with Lag

Weekly:
CNX IT.USDINR.ccf.weekly

Daily:
CNX IT.USDINR.ccf.daily

Conclusion

Currency moves alone cannot be your go-to explanation for fluctuations in tech stocks.

Categories Investing Insight, Your Money Tags myth, quant

Gaps and the Pre-Open Call Auction

shyam May 1, 2015 0

tl;dr

You should not treat opening prices before and after October 18, 2010 the same.

Call Auction in the pre-open session

If you don’t know how the pre-open session works, here’s a good explainer from BSE:

When you run back-tests that use the opening price, this change will most likely trip you up.

Before and after

Nifty opening gaps since 2000:

CNX NIFTY.gap

Notice the shift in the median before and after the auction was introduced (all figures in %):
gap summary

Before:
CNX NIFTY.hist.2

CNX NIFTY.hist.3

After:

CNX NIFTY.hist.4

Conclusion

One way to make the opening prices comparable is to take tick-level data and compute a synthetic opening price yourself, just like how the closing price in computed. And you can use this synthetic open across your entire data set.

Otherwise, you will have to take you back-test results with a healthy dose of skepticism and make sure that there is enough room in your analysis to account for this.

Categories Investing Insight, Your Money Tags quant

Long-Short Trend Following

shyam April 30, 2015 0

Prior Work

We had discussed the SMA On/Off Switch and its ability to escape the worst days. Based on this finding, we setup a Tactical Theme that would go long NIFTYBEES and JUNIORBEES if the CNX 100 index is trading above its 50-day SMA and move into LIQUIDBEES otherwise.

What if, we could go long and short?

Naive Long-Short

Here’s how going long above 50-DMA and short below 50-DMA on the CNX 100 since 2001 compares:

CNX 100.02-Jan-2007.28-Apr-2015.long.short
Long-Short SMA (black), Long-Only SMA (red) and Buy & Hold (green)

It looks like going both long and short is not significantly better than a long-only tactical strategy.

Long-Short with Volatility

But what if, we add a volatility metric into the mix? The logic here is that corrections are preceded by a bout of volatility. So if you go short if either or the volatility signal or the 50-DMA indicates a negative bias and long otherwise:

CNX 100.02-Jan-2007.28-Apr-2015.long.short.volatility
Long-Short SMA w/ Volatility (black), Long-Only SMA w/ Volatility (red), Long-Only SMA (green) and Buy & Hold (blue)

It looks like there is significant alpha in the combination approach.

Long-Short NIFTY and BANKNIFTY

NIFTY returns since 2001:
CNX NIFTY.01-Jan-2001.28-Apr-2015.long.short.volatility

And the same for the BANK NIFTY since 2006:

CNX BANK.12-Jan-2006.28-Apr-2015.long.short.volatility

NIFTY and BANKNIFTY since 2011:

CNX NIFTY.03-Jan-2011.28-Apr-2015.long.short.volatility

CNX BANK.03-Jan-2011.28-Apr-2015.long.short.volatility

NIFTY and BANKNIFTY since 2013:

CNX NIFTY.01-Jan-2013.28-Apr-2015.long.short.volatility

CNX BANK.01-Jan-2013.28-Apr-2015.long.short.volatility
Long-Short Combo (black), Long-Only Combo (red), Long-Only Tactical (green) and Buy & Hold (blue)

Conclusion

It appears that there is long-term alpha in using a combination of volatility and 50-DMA to implement a long-short strategy. To put this to test using real-time data, we have created a theme to make it easy for you to follow along: Trend Long-Short.

Categories Investing Insight, Your Money Tags BANKNIFTY, NIFTY, quant, theme

Analyzing the Analysts

shyam April 29, 2015 0

When we analyzed the price targets of various research analysts for last year, this is what we found:

Most ‘BUY’ ratings were on stocks that had already gone up significantly. The previous 100-day returns before a ‘BUY’ was announced were +23.29% (mean) and +21.49% (median). The next 100-day returns of the same set of stocks were +18.24% (mean) / +17.98% (median.)

For stocks rated ‘SELL’, the previous 100-day returns were +16.93%/+18.47% (mean/median). And they ended up under-performing the ‘BUY’ pool of stocks: +12.50%/+14.14%. But only a brave soul would have gone short in 2014. Even the next 5-day returns for ‘SELL’ rated stocks were +1.22%/0.77%.

One can be excused for thinking that analysts were just chasing momentum, given the above summaries.

BUY’s did not seem to have a dominant short-term effect:
5-day : +0.97%/+0.76%
20-day: +3.78%/+2.90%
50-day: +11.03%/+11.40%

On an average, they did out-perform the CNX 100 index. Next 100-day CNX 100 returns:
BUYs: +12.33%/+11.67%
SELLs: +13.23%/+12.04%

Out of the 718 ratings we analyzed for the year 2014, 565 (~79%) were BUYs and 12.25% were SELLs.

It will be interesting to see how they do this year.

You can download the data here.

Categories Investing Insight, Your Money

Leaders and Laggards

shyam March 24, 2015 0

Lagged correlations

Excess winter snow-fall in the Himalayas lead to floods in Bangladesh during spring. If we know that that there was excess snow-fall in the Himalayas this season, we can be better prepared to handle the floods four-months from now. This is the idea behind studying lagged correlations.

If we took a pair of sector indices and lagged their returns, can we find an index that “leads” an other and profit from it?

The chart above is called the cross-correlation plot. It shows that there are two lags, 5 and 9, where CNX BANK lags CONSUMPTION. A scatter plot shows how monthly-returns are correlated to each other across different lags and confirms the relationship:

Finding

We found a number of index pairs that lead/follow one another. In addition to the CNX BANK and CNX CONSUMPTION indices above, CNX INFRA and CNX CONSUMPTION, CNX IT and CNX FINANCE, CNX CONSUMPTION and AUTO display this dynamic.

Data mining warning

We cannot draw any conclusion from this “finding.” We mined 20 indices over 5 years to dig these nuggets out. The result is spurious. From a statistical point of view, there is no index that consistently leads or lags another.

Categories Investing Insight, Your Money Tags returns