{"id":{"repo_id":"vilnius","oai_identifier":"oai:vu.lt:elaba:308317220"},"canonical_url":"https://search.ndltd.org/etd/vilnius/oai:vu.lt:elaba:308317220","repository":{"repo_id":"vilnius","name":"Vilnius University","base_url":"https://epublications.vu.lt/oai"},"display":{"title":"Comparing ols, fixed effects, and double machine learning for estimating the effect of r&d on firm performance: evidence from eu industrial firms /","abstract":"This thesis compares ordinary least squares, fixed effects and partially linear double machine learning on the effect of R&D on firm performance. This analysis uses a dataset that consists of 13,577 firm-year observations on 1,337 EU industrial firms from the EU Industrial R&D Investment Scoreboard, covering the 2011-2024 period, so that any difference captured in the coefficient could be attributed to the estimator rather than to discrepancies in the data. The whole thesis is organised into three main chapters. Chapter 1 reviews the theoretical and empirical literature on R&D and firm performance as well as the econometric theory surrounding the three model specifications. Chapter 2 describes the data, variable selection, analysis strategy and the diagnostic tests. Chapter 3 presents the descriptive statistics, the OLS, FE, and DML estimates at lags 0, 1, and 2, the comparison across estimators, and the diagnostic test results. Four main findings can be highlighted from this analysis. First, R&D intensity has a negative same-year effect on profitability across all three model specifications. Second, log R&D expenditure has a positive effect on log sales of 0.16 under the DML model and 0.21 under the Fixed Effects model. Third, sales growth displays a small negative same-year effect that turns positive at lags 1 and 2 when R&D is measured as intensity, which is consistent with the knowledge-capital framework of R&D. Fourth, DML estimates remain stable across all three tiers (Small, Medium, Large), suggesting that the results are not sensitive to the choice of controls used. The use of firm and year fixed effects and clustered standard errors is supported by the results of our diagnostic tests. The thesis concludes that comparing DML and FE results on the same data provides more information than only using one preferred estimator. The lagged model specifications should be reported next to the same-year ones, and future empirical research should combine FE with machine-learning models to also analyse the effect of R&D on firm performance in small and medium-sized firms.","abstract_html":"This thesis compares ordinary least squares, fixed effects and partially linear double machine learning on the effect of R&amp;D on firm performance. This analysis uses a dataset that consists of 13,577 firm-year observations on 1,337 EU industrial firms from the EU Industrial R&amp;D Investment Scoreboard, covering the 2011-2024 period, so that any difference captured in the coefficient could be attributed to the estimator rather than to discrepancies in the data. The whole thesis is organised into three main chapters. Chapter 1 reviews the theoretical and empirical literature on R&amp;D and firm performance as well as the econometric theory surrounding the three model specifications. Chapter 2 describes the data, variable selection, analysis strategy and the diagnostic tests. Chapter 3 presents the descriptive statistics, the OLS, FE, and DML estimates at lags 0, 1, and 2, the comparison across estimators, and the diagnostic test results. Four main findings can be highlighted from this analysis. First, R&amp;D intensity has a negative same-year effect on profitability across all three model specifications. Second, log R&amp;D expenditure has a positive effect on log sales of 0.16 under the DML model and 0.21 under the Fixed Effects model. Third, sales growth displays a small negative same-year effect that turns positive at lags 1 and 2 when R&amp;D is measured as intensity, which is consistent with the knowledge-capital framework of R&amp;D. Fourth, DML estimates remain stable across all three tiers (Small, Medium, Large), suggesting that the results are not sensitive to the choice of controls used. The use of firm and year fixed effects and clustered standard errors is supported by the results of our diagnostic tests. The thesis concludes that comparing DML and FE results on the same data provides more information than only using one preferred estimator. The lagged model specifications should be reported next to the same-year ones, and future empirical research should combine FE with machine-learning models to also analyse the effect of R&amp;D on firm performance in small and medium-sized firms.","abstract_has_math":false,"creators":["Jočys, Justinas,"],"institution":"Institutional Repository of Vilnius University","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-19T10:12:55Z","subjects":["R&D, firm performance, double machine learning, fixed effects, ordinary least squares, causal inference, EU Industrial R&D Investment Scoreboard"],"languages":["eng"],"rights":["info:eu-repo/semantics/openAccess"],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://repository.vu.lt/VU:ELABAETD308317220&prefLang=en_US","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://repository.vu.lt/VU:ELABAETD308317220&prefLang=en_US","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":"open","embargo_until":null,"access_note":null},"source_record":{"url":"https://epublications.vu.lt/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Avu.lt%3Aelaba%3A308317220","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Jočys, Justinas,"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026"]},{"key":"dc:publisher","label":"Institution","values":["Institutional Repository of Vilnius University"]},{"key":"dc:relation","label":"Dc Relation","values":["https://epublications.vu.lt/object/elaba:308317220/308317220.pdf"]},{"key":"dc:type","label":"Dc Type","values":["info:eu-repo/semantics/bachelorThesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["R&D, firm performance, double machine learning, fixed effects, ordinary least squares, causal inference, EU Industrial R&D Investment Scoreboard"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["info:eu-repo/semantics/openAccess"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://repository.vu.lt/VU:ELABAETD308317220&prefLang=en_US"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["This thesis compares ordinary least squares, fixed effects and partially linear double machine learning on the effect of R&D on firm performance. This analysis uses a dataset that consists of 13,577 firm-year observations on 1,337 EU industrial firms from the EU Industrial R&D Investment Scoreboard, covering the 2011-2024 period, so that any difference captured in the coefficient could be attributed to the estimator rather than to discrepancies in the data. The whole thesis is organised into three main chapters. Chapter 1 reviews the theoretical and empirical literature on R&D and firm performance as well as the econometric theory surrounding the three model specifications. Chapter 2 describes the data, variable selection, analysis strategy and the diagnostic tests. Chapter 3 presents the descriptive statistics, the OLS, FE, and DML estimates at lags 0, 1, and 2, the comparison across estimators, and the diagnostic test results. Four main findings can be highlighted from this analysis. First, R&D intensity has a negative same-year effect on profitability across all three model specifications. Second, log R&D expenditure has a positive effect on log sales of 0.16 under the DML model and 0.21 under the Fixed Effects model. Third, sales growth displays a small negative same-year effect that turns positive at lags 1 and 2 when R&D is measured as intensity, which is consistent with the knowledge-capital framework of R&D. Fourth, DML estimates remain stable across all three tiers (Small, Medium, Large), suggesting that the results are not sensitive to the choice of controls used. The use of firm and year fixed effects and clustered standard errors is supported by the results of our diagnostic tests. The thesis concludes that comparing DML and FE results on the same data provides more information than only using one preferred estimator. The lagged model specifications should be reported next to the same-year ones, and future empirical research should combine FE with machine-learning models to also analyse the effect of R&D on firm performance in small and medium-sized firms."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Comparing ols, fixed effects, and double machine learning for estimating the effect of r&d on firm performance: evidence from eu industrial firms /","Mažiausių kvadratų, fiksuotų efektų ir dvigubo mašininio mokymosi metodų palyginimas vertinant mokslinių tyrimų ir eksperimentinės plėtros poveikį įmonių veiklos rezultatams: Europos Sąjungos pramonės įmonių duomenys."]}]}],"canonical_facts":{"dc:creator":["Jočys, Justinas,"],"dc:date":["2026"],"dc:description":["This thesis compares ordinary least squares, fixed effects and partially linear double machine learning on the effect of R&D on firm performance. This analysis uses a dataset that consists of 13,577 firm-year observations on 1,337 EU industrial firms from the EU Industrial R&D Investment Scoreboard, covering the 2011-2024 period, so that any difference captured in the coefficient could be attributed to the estimator rather than to discrepancies in the data. The whole thesis is organised into three main chapters. Chapter 1 reviews the theoretical and empirical literature on R&D and firm performance as well as the econometric theory surrounding the three model specifications. Chapter 2 describes the data, variable selection, analysis strategy and the diagnostic tests. Chapter 3 presents the descriptive statistics, the OLS, FE, and DML estimates at lags 0, 1, and 2, the comparison across estimators, and the diagnostic test results. Four main findings can be highlighted from this analysis. First, R&D intensity has a negative same-year effect on profitability across all three model specifications. Second, log R&D expenditure has a positive effect on log sales of 0.16 under the DML model and 0.21 under the Fixed Effects model. Third, sales growth displays a small negative same-year effect that turns positive at lags 1 and 2 when R&D is measured as intensity, which is consistent with the knowledge-capital framework of R&D. Fourth, DML estimates remain stable across all three tiers (Small, Medium, Large), suggesting that the results are not sensitive to the choice of controls used. The use of firm and year fixed effects and clustered standard errors is supported by the results of our diagnostic tests. The thesis concludes that comparing DML and FE results on the same data provides more information than only using one preferred estimator. The lagged model specifications should be reported next to the same-year ones, and future empirical research should combine FE with machine-learning models to also analyse the effect of R&D on firm performance in small and medium-sized firms."],"dc:format":["application/pdf"],"dc:identifier":["https://repository.vu.lt/VU:ELABAETD308317220&prefLang=en_US"],"dc:language":["eng"],"dc:publisher":["Institutional Repository of Vilnius University"],"dc:relation":["https://epublications.vu.lt/object/elaba:308317220/308317220.pdf"],"dc:rights":["info:eu-repo/semantics/openAccess"],"dc:subject":["R&D, firm performance, double machine learning, fixed effects, ordinary least squares, causal inference, EU Industrial R&D Investment Scoreboard"],"dc:title":["Comparing ols, fixed effects, and double machine learning for estimating the effect of r&d on firm performance: evidence from eu industrial firms /","Mažiausių kvadratų, fiksuotų efektų ir dvigubo mašininio mokymosi metodų palyginimas vertinant mokslinių tyrimų ir eksperimentinės plėtros poveikį įmonių veiklos rezultatams: Europos Sąjungos pramonės įmonių duomenys."],"dc:type":["info:eu-repo/semantics/bachelorThesis"]},"updated_at":"2026-11-19T10:12:55Z"}