Improving External Validity of Machine Learning, Reduced Form, and Structural Macroeconomic Models using Panel Data

2021 ◽  
Author(s):  
Cameron Fen ◽  
Samir Undavia
Keyword(s):  
Machine Learning ◽  
Panel Data ◽  
External Validity ◽  
Reduced Form ◽  
Macroeconomic Models

Common time variation of parameters in reduced-form macroeconomic models

2015 ◽  
Vol 0 (0) ◽  
Author(s):  
Dalibor Stevanovic
Keyword(s):  
Time Variation ◽  
Reduced Form ◽  
Time Varying ◽  
Vector Autoregressions ◽  
Macroeconomic Models ◽  
Variation Of Parameters ◽  
Common Time ◽  
Time Varying Parameter ◽  
Low Dimension ◽  
Structural Vector

AbstractStandard time varying parameter (TVP) models usually assume independent stochastic processes. In this paper, I show that the number of underlying sources of parameters’ time variation is likely to be small, and provide empirical evidence for factor structure amongst TVPs of popular macroeconomic models. In order to test for the presence of low dimension sources of time variation in parameters and estimate their magnitudes, I develop the factor time varying parameter (Factor-TVP) framework and apply it to [Primiceri, G.E. (2005), “Time Varying Structural Vector Autoregressions and Monetary Policy,”

scholarly journals Machine learning algorithm to predict anterior cruciate ligament revision demonstrates external validity

2022 ◽  
Author(s):  
R. Kyle Martin ◽  
Solvejg Wastvedt ◽  
Ayoosh Pareek ◽  
Andreas Persson ◽  
Håvard Visnes ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acl Reconstruction ◽  
External Validity ◽  
Learning Algorithm ◽  
Cruciate Ligament ◽  
External Validation ◽  
Model Performance ◽  
Knee Ligament ◽  
Graft Choice ◽  
Anterior Cruciate

Abstract Purpose External validation of machine learning predictive models is achieved through evaluation of model performance on different groups of patients than were used for algorithm development. This important step is uncommonly performed, inhibiting clinical translation of newly developed models. Machine learning analysis of the Norwegian Knee Ligament Register (NKLR) recently led to the development of a tool capable of estimating the risk of anterior cruciate ligament (ACL) revision (https://swastvedt.shinyapps.io/calculator_rev/). The purpose of this study was to determine the external validity of the NKLR model by assessing algorithm performance when applied to patients from the Danish Knee Ligament Registry (DKLR). Methods The primary outcome measure of the NKLR model was probability of revision ACL reconstruction within 1, 2, and/or 5 years. For external validation, all DKLR patients with complete data for the five variables required for NKLR prediction were included. The five variables included graft choice, femur fixation device, KOOS QOL score at surgery, years from injury to surgery, and age at surgery. Predicted revision probabilities were calculated for all DKLR patients. The model performance was assessed using the same metrics as the NKLR study: concordance and calibration. Results In total, 10,922 DKLR patients were included for analysis. Average follow-up time or time-to-revision was 8.4 (± 4.3) years and overall revision rate was 6.9%. Surgical technique trends (i.e., graft choice and fixation devices) and injury characteristics (i.e., concomitant meniscus and cartilage pathology) were dissimilar between registries. The model produced similar concordance when applied to the DKLR population compared to the original NKLR test data (DKLR: 0.68; NKLR: 0.68–0.69). Calibration was poorer for the DKLR population at one and five years post primary surgery but similar to the NKLR at two years. Conclusion The NKLR machine learning algorithm demonstrated similar performance when applied to patients from the DKLR, suggesting that it is valid for application outside of the initial patient population. This represents the first machine learning model for predicting revision ACL reconstruction that has been externally validated. Clinicians can use this in-clinic calculator to estimate revision risk at a patient specific level when discussing outcome expectations pre-operatively. While encouraging, it should be noted that the performance of the model on patients undergoing ACL reconstruction outside of Scandinavia remains unknown. Level of evidence III.

scholarly journals The Employment in Innovative Enterprises in Europe

2022 ◽  
Author(s):  
Lucio Laureti ◽  
Costantiello Alberto ◽  
Marco Maria Matarrese ◽  
Angelo Leogrande
Keyword(s):  
Machine Learning ◽  
Panel Data ◽  
Fixed Effects ◽  
Value Added ◽  
Machine Learning Algorithms ◽  
High Tech ◽  
Dynamic Panel ◽  
Silhouette Coefficient ◽  
High Tech Product ◽  
And Training

Abstract In this article we evaluate the determinants of the Employment in Innovative Enterprises in Europe. We use data from the European Innovation Scoreboard of the European Commission for 36 countries in the period 2000-2019 with Panel Data with Fixed Effects, Panel Data with Random Effects, Dynamic Panel, WLS and Pooled OLS. We found that the “Employment in Innovative Enterprises in Europe” is positively associated with “Broadband Penetration in Europe”, “Foreign Controlled Enterprises Share of Value Added”, “Innovation Index”, “Medium and High-Tech Product Exports” and negatively associated to “Basic School Entrepreneurial Education and Training”, “International Co-Publications”, and “Marketing or Organizational Innovators”. Secondly, we perform a cluster analysis with the k-Means algorithm optimized with the Silhouette Coefficient and we found the presence of four different clusters. Finally, we perform a comparison among eight different machine learning algorithms to predict the level of “Employment in Innovative Enterprises” in Europe and we found that the Linear Regression is the best predictor.

scholarly journals The Determinants of Design Applications in Europe

2021 ◽  
Author(s):  
Angelo Leogrande ◽  
Alberto Costantiello ◽  
Lucio Laureti ◽  
Domenico Leogrande
Keyword(s):  
Machine Learning ◽  
Panel Data ◽  
Fixed Effects ◽  
Econometric Models ◽  
Machine Learning Algorithms ◽  
Dynamic Panel ◽  
A Value ◽  
Silhouette Coefficient ◽  
Firm Investments ◽  
Intellectual Assets

Abstract In this article we estimate the level of “Design Application” in 37 European Countries in the period 2010-2019. We use data from the European Innovation Scoreboard-EIS of the European Commission. We perform four econometric models i.e., Pooled OLS, Panel Data with Random Effects, Panel Data with Fixed Effects, Dynamic Panel. We found that the level of Design Applications is negatively associated to “Enterprise Births”, “Finance and Support”, “Firm Investments” and positively associated with “Venture Capital”, “Turnover share large enterprises”, “R&D expenditure public sector”, “Intellectual Assets”. In adjunct we perform a cluster analysis with the application of the k-Means algorithm optimized with the Silhouette Coefficient and we found three different clusters. Finally, we confront eight different machine learning algorithms to predict the level of “Design Application” and we found that the Tree Ensemble is the best predictor with a value for the 30% of the dataset analyzed that is expected to decrease in mean of -12,86%.

Application of the Menges model in a modern economy based on panel data from the Central District of the Russian Federation

2021 ◽  
Author(s):  
Ekaterina Dmitrievna Myagotina ◽  
Ilona Vladimirovna Tregub
Keyword(s):  
Panel Data ◽  
Russian Federation ◽  
Industrial Production ◽  
Research Work ◽  
National Income ◽  
Federal District ◽  
Statistical Information ◽  
Reduced Form ◽  
Central Federal District ◽  
The Russian Federation

Nowadays analysis of various econometric models is used to study a significant amount of updated statistical information and find out relationships between statistical economic indicators, that were not investigated earlier. Main idea of investigation in this research work — to find out whether factors influence on the GRP, Consumption, Profit of the organizations and Investments or the model is outdated and non-applicable nowadays. The results obtained in the framework of this research will give us understanding changing which factors (exogeneous variables) — for example, increasing or decreasing credit and deposit rates, tax rates will rise national income, consumer expenditure, investment and operating surplus and, as a result, will accelerate economic development of the Russian Federation. The purpose is to evaluate whether the Menges model which includes all the indicators mentioned as dependent variables is applicable in the modern conditions of the Central Federal District in Russia or not and to find out whether there are other factors which also have an impact on endogenous variables. The object — a set of the panel data of economic statistical information of the Central Federal District in Russia (2008–2013). The subject — the reduced form of Menges model including GRP, Consumption, Profit of the organizations and Net Investments. According to our research, it is reasonable enough to increase the volume of industrial production in order to increase GRP both economically and with the help of the Menges econometric model. Besides that, it is also reasonable to increase the volume of industrial production in order to get a higher cost of investment.

Identifying reduced-form relations with panel data: The case of pollution and income

2009 ◽  
Vol 58 (1) ◽  
pp. 27-42 ◽  
Author(s):  
Herman R.J. Vollebergh ◽  
Bertrand Melenberg ◽  
Elbert Dijkgraaf
Keyword(s):  
Panel Data ◽  
Reduced Form
Sign in / Sign up

Export Citation Format

Share Document