Replication Materials - Assumption Errors and Forecast Accuracy: A Partial Linear Instrumental Variable and Double Machine Learning Approach

Fuente: Zenodo
Saved in:
Bibliographic Details
Main Authors: Schult, Christoph, Heinisch, Katja, Scaramella, Fabio
Format: Recurso digital
Published: Zenodo 2025
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866901415810564096
author Schult, Christoph
Heinisch, Katja
Scaramella, Fabio
author_facet Schult, Christoph
Heinisch, Katja
Scaramella, Fabio
contents <div> <p>This repository contains the replication files for the paper <em>“Assumption Errors and Forecast Accuracy: A Partial Linear Instrumental Variable and Double Machine Learning Approach”</em>. The study investigates how conditioning variable forecast errors (oil prices, world trade, exchange rates) affect GDP forecast accuracy, using a novel econometric framework based on Partial Linear Instrumental Variables (PLIV) and Double Machine Learning (DML).</p> <p>The replication package includes:</p> <ul> <li> <p><strong>Data</strong>: Forecast and conditioning assumption datasets (1992–2019) covering German GDP forecasts from major national and international institutions.</p> </li> <li> <p><strong>Code</strong>: R scripts implementing OLS, 2SLS, and PLIV-DML estimations using the <code>DoubleML</code> and <code>mlr3</code> packages.</p> </li> <li> <p><strong>Figures and Tables</strong>: Scripts to reproduce all figures and tables reported in the manuscript (main text and appendix).</p> </li> <li> <p><strong>Instructions</strong>: A README file describing how to run the code, install dependencies, and reproduce results.</p> </li> </ul> </div>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_17035188
institution Zenodo
language
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle Replication Materials - Assumption Errors and Forecast Accuracy: A Partial Linear Instrumental Variable and Double Machine Learning Approach
Schult, Christoph
Heinisch, Katja
Scaramella, Fabio
<div> <p>This repository contains the replication files for the paper <em>“Assumption Errors and Forecast Accuracy: A Partial Linear Instrumental Variable and Double Machine Learning Approach”</em>. The study investigates how conditioning variable forecast errors (oil prices, world trade, exchange rates) affect GDP forecast accuracy, using a novel econometric framework based on Partial Linear Instrumental Variables (PLIV) and Double Machine Learning (DML).</p> <p>The replication package includes:</p> <ul> <li> <p><strong>Data</strong>: Forecast and conditioning assumption datasets (1992–2019) covering German GDP forecasts from major national and international institutions.</p> </li> <li> <p><strong>Code</strong>: R scripts implementing OLS, 2SLS, and PLIV-DML estimations using the <code>DoubleML</code> and <code>mlr3</code> packages.</p> </li> <li> <p><strong>Figures and Tables</strong>: Scripts to reproduce all figures and tables reported in the manuscript (main text and appendix).</p> </li> <li> <p><strong>Instructions</strong>: A README file describing how to run the code, install dependencies, and reproduce results.</p> </li> </ul> </div>
title Replication Materials - Assumption Errors and Forecast Accuracy: A Partial Linear Instrumental Variable and Double Machine Learning Approach
url https://doi.org/10.5281/zenodo.17035188