EconBase
← All papers

Dual Interpretation of Machine Learning Forecasts

Philippe Goulet Coulombe, Maximilian Goebel, Karin Klieber

arXiv 17 Dec 2024 · Econometrics · 4 citations (OpenAlex)

arXiv:2412.13076 · PDF · DOI · OpenAlex · Extracted main text

Abstract

Machine learning predictions are typically interpreted as the sum of contributions of predictors. Yet, each out-of-sample prediction can also be expressed as a linear combination of in-sample values of the predicted variable, with weights corresponding to pairwise proximity scores between current and past economic events. While this dual route leads nowhere in some contexts (e.g., large cross-sectional datasets), it provides sparser interpretations in settings with many regressors and little training data-like macroeconomic forecasting. In this case, the sequence of contributions can be visualized as a time series, allowing analysts to explain predictions as quantifiable combinations of historical analogies. Moreover, the weights can be viewed as those of a data portfolio, inspiring new diagnostic measures such as forecast concentration, short position, and turnover. We show how weights can be retrieved seamlessly for (kernel) ridge regression, random forest, boosted trees, and neural networks. Then, we apply these tools to analyze post-pandemic forecasts of inflation, GDP growth, and recession probabilities. In all cases, the approach opens the black box from a new angle and demonstrates how machine learning models leverage history partly repeating itself.

Citation extraction

88
references
117
in-text mentions
88
distinct cited
2
self-citations
20,481
main-text words

appendix boundary found by appendix_command · 92% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Goulet Coulombe, P (2024) A neural phillips curve and a deep output gap0.9098475%
2Geertsema, P. and Lu, H (2023) Instance-based Explanations for Gradient Boosting Machine Predictions with AXIL Weights0.87472100%
3Goulet Coulombe, P (2024) The macroeconomy as a random forest0.81142100%
4Ghorbani, A. and Zou, J (2019) Data shapley: Equitable valuation of data for machine learning0.73732100%
5Koster, N. and Krüger, F (2024) Simplifying random forests' probabilistic forecasts0.73732100%
6McCracken, M. W. and Ng, S (2016) Fred-md: A monthly database for macroeconomic research0.73732100%
7Buckmann, M. and Joseph, A (2023) An interpretable machine learning workflow with an application to economic forecasting0.64422100%
8Dendramis, Y., Kapetanios, G., and Marcellino, M (2020) A similarity-based approach for macroeconomic forecasting0.64422100%
9Foroni, C., Marcellino, M., and Stevanovic, D (2022) Forecasting the covid-19 recession and recovery: Lessons from the financial crisis0.64422100%
10Lin, Y. and Jeon, Y (2006) Random forests and adaptive nearest neighbors0.64422100%

Showing the top 10 of 88 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
10.29cm 22.5524 dpd Ordinary Least Squares as an Attention Mechanism . 0.25cm0.84333
20.5cm dpd LGB+: A Macroeconomic Forecasting Road Test . 0.25cm0.84333
30.09cm 24.9522 dpd Opening the Black Box of Local Projections . 0.4cm0.81142
41.2cm 2222dpd Quantifying the Risk–Return Tradeoff in Forecasting 1.65cm0.64422
50.9cm 18.9522 dpd An Adaptive Moving Average for Macroeconomic Monitoring . 0.25cm0.40511
6Who Saw It Coming? Historical Experience and the 2021 Inflation Forecast Failure0.40511