Abstract
We use machine learning methods to examine the power of Treasury term spreads and other financial market and macroeconomic variables to forecast US recessions, vis-à-vis probit regression. In particular we propose a novel strategy for conducting cross-validation on classifiers trained with macro/financial panel data of low frequency and compare the results to those obtained from standard k-folds cross-validation. Consistent with the existing literature we find that, in the time series setting, forecast accuracy estimates derived from k-folds are biased optimistically, and cross-validation strategies which eliminate data "peeking" produce lower, and perhaps more realistic, estimates of forecast accuracy. More strikingly, we also document rank reversal of probit, Random Forest, XGBoost, LightGBM, neural network and support-vector machine classifier forecast performance over the two cross-validation methodologies. That is, while a k-folds cross-validation indicates tha t the forecast accuracy of tree methods dominates that of neural networks, which in turn dominates that of probit regression, the more conservative cross-validation strategy we propose indicates the exact opposite, and that probit regression should be preferred over machine learning methods, at least in the context of the present problem. This latter result stands in contrast to a growing body of literature demonstrating that machine learning methods outperform many alternative classification algorithms and we discuss some possible reasons for our result. We also discuss techniques for conducting statistical inference on machine learning classifiers using Cochrane's Q and McNemar's tests; and use the SHapley Additive exPlanations (SHAP) framework to decompose US recession forecasts and analyze feature importance across business cycles.
Abstract
In this paper, we investigate the forecasting ability of the yield curve in terms of the U.S. real GDP cycle. More specifically, within a Machine Learning (ML) framework, we use data from a variety of short (treasury bills) and long term interest rates (bonds) for the period from 1976:Q3 to 2011:Q4 in conjunction with the real GDP for the same period, to create a model that can successfully forecast output fluctuations (inflation and output gaps) around its long-run trend. We focus our attention in correctly forecasting the instances of output gaps referred for the purposes of our analysis here as recessions. In this effort, we applied a Support Vector Machines (SVM) technique for classification. The results show that we can achieve an overall forecasting accuracy of 66,7% and a 100% accuracy in forecasting recessions.
Abstract
We develop early warning models for financial crisis prediction by applying machine learning techniques to macrofinancial data for 17 countries over 1870–2016. Most nonlin-ear machine learning models outperform logistic regression in out-of-sample predictions and forecasting. We identify economic drivers of our machine learning models using a novel framework based on Shapley values, uncovering nonlinear relationships between the predic-tors and crisis risk. Throughout, the most important predictors are credit growth and the slope of the yield curve, both domestically and globally. A flat or inverted yield curve is of most concern when nominal interest rates are low and credit growth is high.