Machine learning in macroeconomic forecasting
Authors
Date
2022Copyright
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Dataa on aina ollut saatavilla paljon taloudesta, mutta sen kaiken käyttäminen talouden ennustamisessa on ollut hankalaa. Perinteiset ennustamisen ja arvioinnin mallit eivät ole osoittautuneet olevan kovin tarkkoja makrotalouden ennustamisessa. Modernit koneoppimisen menetelmät ovat osoittautuneet hyviksi monessa eri tilanteessa ja monella eri alalla. Koneoppiminen on vahvimmillaan juuri ennusteiden tekemisessä. Taloutta on aina pyritty ennustamaan ekonometrisillä malleilla, mutta koneoppimisen on huomattu monessa paikassa olevan tarkempi ennusteissaan kuin perinteisemmät mallit. Koneoppimista voidaan käyttää työkaluna ennustamisessa monien eri metodien ja algoritmien kautta, joilla kaikilla on omat vahvuutensa sekä heikkoutensa. Jokaista näistä voidaan käyttää erilaisten ennusteiden tekemisessä juuri niiden vahvuuksien ja heikkouksien perusteella. Ennustaa voi esimerkiksi bruttokansantuotteen kasvua ja pienenemistä, inflaatiota tai velkakirjojen korkoja. Koneoppimisen menetelmien on huomattu olevan tehokkaampia kuin perinteisten aikasarja-analyysien, ja vain tulevaisuus näyttää kuinka tarkasti koneoppimista opitaan hyödyntämään makrotalouden ennustamisessa. Tämä kirjallisuuskatsaus avaa koneoppimista, sekä perehtyy tarkemmin sen eri metodeihin ja kertoo miten koneoppimista ja näitä eri metodeja voidaan käyttää talouden ennustamisessa.
...
There has always been large amounts of data available of the economy but using all of this to make predictions about the economy has been difficult. Traditional models used in forecasting and in estimates have not proven to be that accurate. The modern methods that machine learning provides have proven to perform well in many different situations and in many different disciplines. Machine learning is at its strongest in making predictions. Econometric models have always tried to forecast the economy, but it has been noted that machine learning is more accurate in its predictions than the more traditional models. Machine learning can be used as a tool in forecasting through many different methods and algorithms which all have their individual strengths and weaknesses. Each of these can be used in making different kinds of predictions based on their strengths and weaknesses. Some good indicators to forecast would be for example the falls and rises of GDP, inflation, or bonds’ interest rates. Machine learning has already been proven to be more efficient than time-series analysis and only the future will tell how well the macroeconomy will be forecasted with machine learning. This literature review explains what machine learning is, familiarizes the reader with different machine learning methods, and explains how machine learning and its methods can be used in economic forecasting.
...
Keywords
Metadata
Show full item recordCollections
- Kandidaatintutkielmat [5334]
Related items
Showing items with similar title or keywords.
-
Forecasting real economic activity and inflation : the role of geopolitical risks and economic policy uncertainty in major economies
Hundal, Amaya (2023)The increasing prominence of geopolitical risks and economic policy uncertainties globally has brought their impacts on overall economies to the forefront in both academia and policymaking. Current literature, while being ... -
Economic policy uncertainty effects for forecasting future real economic activity
Junttila, Juha-Pekka; Vataja, Juuso (Elsevier B.V., 2018)Recently introduced measures for Economic Policy Uncertainty (EPU) included in the data from 1997 - 2016 have a role in forecasting out-of-sample values for the future real economic activity for both the euro area and ... -
Domain-specific transfer learning in the automated scoring of tumor-stroma ratio from histopathological images of colorectal cancer
Petäinen, Liisa; Väyrynen, Juha P.; Ruusuvuori, Pekka; Pölönen, Ilkka; Äyrämö, Sami; Kuopio, Teijo (Public Library of Science (PLoS), 2023)Tumor-stroma ratio (TSR) is a prognostic factor for many types of solid tumors. In this study, we propose a method for automated estimation of TSR from histopathological images of colorectal cancer. The method is based on ... -
Comparing the forecasting performance of logistic regression and random forest models in criminal recidivism
Aaltonen, Olli-Pekka (2016)Rikosseuraamusalalla on viime vuosina kehitetty uusintarikollisuutta ennustavia malleja (Tyni, 2015), jotka perustuvat tyypillisesti rekisteripohjaisiin mittareihin, jotka mittaavat mm. tuomitun sukupuolta, ikää, rikostaustaa ... -
Predicting Children's Myopia Risk : A Monte Carlo Approach to Compare the Performance of Machine Learning Models
Artiemjew, Piotr; Cybulski, Radosław; Emamian, Mohammad; Grzybowski, Andrzej; Jankowski, Andrzej; Lanca, Carla; Mehravaran, Shiva; Młyński, Marcin; Morawski, Cezary; Nordhausen, Klaus; Pärssinen, Olavi; Ropiak, Krzysztof (SCITEPRESS Science and Technology Publications, 2024)This study presents the initial results of the Myopia Risk Calculator (MRC) Consortium, introducing an innovative approach to predict myopia risk by using trustworthy machine-learning models. The dataset included approximately ...