Economics

More factors, more variance...explained

Risk factor models are at the core of quantitative investing. We’ve been exploring their application within our portfolio series to see if we could create such a model to quantify risk better than using a simplistic volatility measure. That is, given our four portfolios (Satisfactory, Naive, Max Sharpe, and Max Return) can we identify a set of factors that explain each portfolio’s variance relatively well? In our first investigation, we used the classic Fama-French (F-F) three factor model plus momentum.

Tens and twos

Only three months ago, market pundits were getting lathered up about the potential for an inverted yield curve. We discussed that in our post Fed up. But a lot has changed since then. One oft-used measure of the yield curve, the time spread (10-year Treasury yields less 3-month yields), has inverted (gone negative). The NY Fed’s yield curve model sets the probability of recession 12-months hence above 31%, up from over 27% in May.

Fed up

Yield curve predictions are hitting the headlines again here, here, and here, though they’re not quite front page. The alarm bells are ringing since the probability of a recession appears to have increased meaningfully in the past few months. We look at the data to try to infer whether a recession is around the corner. So what is it that has ruffled everyone’s feathers? The NY Fed’s yield curve model estimates the probability of a recession in the next twelve months at an exceedingly precise 27.

What do machines know about the yield curve?

A recession a year from now? As we saw in the last post, when we run a model with a 6-month look forward, it does a fairly reasonable job in predicting a recession, assuming we use a threshold closer to recession base rate. In this post, we look at 12-month look forward and then use the best of the two look forward models to test it on out-of-sample data.

Yield curve predictions twist my noodle

Where did we go wrong? Not another model! As we saw in the last post, one iteration of the yield curve – the spread between 10-year and 3-month Treasuries – doesn’t generate a great model of recession probabilities. Part of this is that recessions are not that common, so we’re trying to find the veritable needle. Another problem is picking the right threshold to say the model is prediciting a greater likelihood of the economy being recession.

Yield curve predictions are really hard

All models are wrong Build the model In the last post, we discussed the yield curve, why investors focus on it, and looked at one measure of the curve – the spread between 10-year and 3-month Treasury yields. In this post, we build a model that tries to quantify the probability that the economy is in recession based on the 10-year/3-month spread. All models are wrong A first question to ask is what kind of model should we build?

Yield curve predictions are hard

Introduction What is a yield curve? (Skip if you already know!) Show me the data! (Start here if don’t want the background) Investing pundits like to quote the yield curve as a nearly infallible indicator for the next recession. But what do the data say? And which yield curve should you use? In this multi-part series we try to answer these questions in as straightforward (though not necessarily simple) a manner as possible.