Sentiment Indicators - Ignore the Singapore Purchasing Manufacturers’ Index - Student's Post

For a mere sentiment measure, the PMI is widely watched by economists and stock people for some strange unknown reason. Here is the SIPMM’s self-description of the PMI, with absolutely no sense of irony:

The Singapore PMI has become a key barometer of the Singapore manufacturing economy and has been highly sought after by local and international news agencies, banks, investment and stock-broking firms.  The Singapore PMI is published monthly in the major language press media, such as The Straits Times, The Business Times, The New Paper, Chinese Daily News LianHe ZaoBao and the Malay Daily News Berita Harian.  The index is also reported regularly in the international media, as well as economic and research agencies worldwide.  In addition, the Channel News Asia and News Radio 93.8 also broadcast the monthly news release of the index.

Any index that is pleased to be reported by The New Paper must have some serious psychological issues going on. Here is my take on the PMI:

(Please switch the t-1 and t+1 on the above graph as that makes more intuitive sense.)
If you squint you can see that the explanatory power of the PMI on the STI is no more than 10 basis points as a coincident indicator, and 15 bps as a leading indicator. On the other hand, the PMI does fabulously as a lagging indicator, explaining 112 bps. By means of comparison, the explanatory power of last month’s STI return on this month is 44 bps.*
Full data is uploaded at the usual place.**
The fact that all these media outlets report this essentially useless data exemplifies the difference between “news” and “news you can use”. We are looking for tomorrow’s weather forecast, but many of the radio stations we tune in to simply talk about yesterday’s weather. ‘Nuff said.
*While the 112 vs the 44 seems strange, this finding is consistent with the human experience that it is much easier to explain the past looking backward (PMI) than it is to explain the future looking backward (STI return t-1).
**By comparison, HK’s PMI does not even keep a historical record and is released only infrequently according to Bloomberg. A sorry state of affairs for sentiment indicators in general.

Sentiment Indicators - The Singapore PMI and ST Sector Indices - Student's Post

Armed with sector information from the FTSE, I have followed up on the prior post looking at the PMI and the STI. The PMI is a sentiment index of purchasers in the manufacturing industry, so I have identified the relevant sector indices to be the FSTBM, FSTIN, and FSTCG, corresponding to the Basic Materials, Industrials, and Consumer Goods indices. I then ran a simple regression on their returns, much like the prior post:

We can see that the PMI doesn’t do that great as a coincident indicator on any of the relevant sectors; the best we get is 34 basis points, and as mentioned in the prior post you can get 44 bps just wagering that this month’s returns are the same as last month’s. In other words, even if you somehow managed to get insider info on the PMI or forecasted the PMI accurately (also suspect; I have recorded PMI survey data on my data subsite for your perusal), you would not be likely at all to make much alpha.
Nothing to write home (or here) about.
However, great news is about “man bites dog”, and a little messing around with the lags show this:
This shows the same regression run on the PMI lagged by 1 month. The R-squared on all three sectors leaps by 7x, 4x, and 9x respectively – in particular, the previous month’s PMI seems to explain almost 3% of the FSTCG Index’s movements over this past decade (N=135). This is a magic number – some quants say that “People have gotten rich off of a 2% r-squared” – however an adult user of statistics would not get overly excited about this if only because the sample size is still not that impressive.
I know some of this talk about R-squared puts some people off, so here’s a simple trading rule conclusion you can make from this study:
When the PMI minus 50 was lower than -0.4 (purchasing managers were bearish), the FSTCG moved up 2.59% in the next month on average. When the PMI minus 50 was above 2.1 (bullish), Consumer Goods stocks dropped on average. The current PMI is at 50.7.

Ratio Reversion - Student's Post

Quote of the Day: “True leadership must be for the benefit of the followers, not the enrichment of the leaders.”
Today’s statistical phenomenon has many names – mean reversion, statistical arbitrage, lead-lag structures – but the trading strategy remains the same. We are leaving the Market Timing family of strategies and going further into the deep end of Mean Reversion for this one. Observe this chart:


We see that once we normalize as of a certain date, the Straits Times Index and the Hang Seng Index track each other closely despite their vastly different numerical levels. Plotting the cumulative degree of STI outperformance results in the green line, which has no further purpose than to show that the ratio of FSSTI to HSI hovers pretty much around the level it is at as of 3 Jan 2001 (this ratio is 7.72). Thus, when the FSSTI outperforms the HSI on any particular day, it is likely to underperform the HSI at some future point (not necessarily the following day). With a long/short market neutral bet we can profit from the expected narrowing of this gap, but there are an infinite number of ways to structure your decision rule. The simplest, in my view, is to bet on the ratio reverting back to 7.72, and to keep at it until the ratio is reestablished (Static Ratio Reversion):

This chart is on a log scale so I have added +1 to all values on the green line in order to display it meaningfully. The purple line shows the profits resulting from the Static Ratio Reversion strategy, and although the end result of 381% profit is impressive, it is worthwhile to note that from 2001 to late 2007 the strategy didn’t have any profit to show for all its sophistication. Also, eyeballing the green and purple lines, it is clear that the strategy only makes money when the green line heads toward the 1.000 line, which marks the 7.72 ratio established on 3rd Jan 2001. This highlights the chief difficulty with the Static Ratio Reversion strategy, as it is a strategy that says that the ratio of FSSTI and HSI on any given day is wrong, but the ratio on one particular day (in this one, 3rd Jan 2001), is “right”. Further, it doesn’t allow for any fundamental movement over time in the ratio itself.
The Moving Average Ratio Reversion method solves this by trading based on reversion to a moving average of past ratios, and not one past ratio on one particular day. This can help anticipate movements in the trend far better, as it can be seen that the 7.72 point is quite often off the mark:

With this in mind, is a slower moving average better or a faster moving average? I don’t have the time or presence of mind to derive the relationship between the predictive ability of the moving average of a ratio and the profit from a ratio reversion strategy on that ratio, but empirically we can observe a few data points and make some inferences:




It seems that the mid 2000s loss periods are avoided with longer MA estimations. I need to do further work on this and don’t have the time, but I leave you with the risk/return statistics:

Index Component Mean Reversion - Student's Post

Citation of the Day: What Happened to the Quants in August 2007?
It’s a holiday so this is a short post. I have completed preliminary work adapting Andrew Lo’s mean reversion strategy to the Straits Times Index, and it looks like this:

It’s looking pretty good, making 4% over a year with a max 2% drawdown UNLEVERED and MARKET NEUTRAL. More analysis to be done, but in the meantime, HAPPY CHRISTMAS!

The Economist Chimes in on Momentum - Students Post

I think I’m on a roll with this media thing. Today’s Economist has an article out on momentum investing that is probably worth your time:

It is also given a mention in the venerable Leader section.
However, whoever wrote the article has in my opinion a very poor understanding of proper momentum investing:

A second puzzle is why the effect has not been arbitraged away. The answer probably lies in timing. Clearly the momentum effect cannot last for ever or share prices would head for infinity. Over long periods (more than three years or so) an opposite anomaly known as the value effect occurs: shares that are depressed in price tend to rebound. Momentum-chasing investors may get caught out by the switch from one effect to the other, especially when they have used borrowed money to try to enhance returns.

This is an unforgivably stupid reason even by their chosen strategy.

Investors who buy the best-performing shares over the previous year earn much higher returns (ten percentage points a year) than those who buy the laggards of the preceding 12 months.

By definition momentum investors cannot be “caught out” by the switch to the 3-year value effect if their momentum horizon is 12 months.
This Student suggests the eggheads at the Economist read up on Cliff Asness’ seminal Value and Momentum Everywhere for a better way to compare value and momentum directly, and for some more informed ruminations on more likely sources of momentum profits.
in my view
The easiest explanation for why momentum investing continues to exist is that it is the one market inefficiency that is impossible to arbitrage. Typical arbitrage involves some conception of the market price being different from fair value, and the trade profits from the market price moving in the opposite direction of the pricing discrepancy. If momentum is real, fair value is a function of market price and the trade profits from the market price moving in the SAME direction of the pricing discrepancy. Attempting to “arbitrage” the momentum effect makes it stronger, not weaker, unlike in all other arbitrage scenarios.
One further comment on momentum: This has obvious links to Soros’ reflexivity, but at least he grounds it in the effect that accessing the capital markets can have on the fundamental fair value of a company. True momentum doesn’t even wait for the stock price to affect the company’s fundamentals; the stock price affects the stock price, and that is that.