Browsing by master's degree program "Master 's Programme in Economics"
Now showing items 1-3 of 3
-
(2019)Prices of different products are followed by statistical offices in order to produce price indices. The quality of products is constantly changing due to creative destruction. When a product leaves market, its price is computed with a method called imputation. Recent studies in United States and France have found that use of imputation may lead to upward bias in inflation. Since price indices are used as deflators when calculating economic growth, such a bias would mean that some of the growth is missed. The aim of this thesis is to study whether such a bias exists in Finland and how large it is. In addition, the channels of innovation induced growth are studied in order to determine from where the potentially missed growth originates. Creative destruction has been incorporated into economic growth models in the early 1990s. In its centre, are firms at the microlevel that innovate and create new products and improve existing ones. It has been shown that it is a key element when economic growth is concerned. New products and improving quality of old varieties is, however, widely recognised problem for price indices. Sources of bias for price statistics has been studied a lot and the changing quality of products is one of the greatest of them. This thesis contributes to this field by recognising a new possible source of bias and its magnitude in Finnish economy. The model used in this thesis is from 2017 paper by Aghion, Bergeaud, Boppart, Klenow and Li. The model is a new keynesian DSGE model with exogenous innovation and it provides an accounting framework which enables the quantification of missing growth. The missing growth is estimated using a so-called market share approach, where market shares of incumbent and entrant producers are exploited to quantify the share of growth that is missed yearly. Another method, namely indirect inference, relies on simulation of the economic growth model. It infers the arrival rates and step sizes of different types of innovations: incumbent of innovation, creative destruction and new product varieties. The simulation also enables for finding the contributions of those innovation types for the economic growth. The contributions provide information on from which type of innovation the majority of growth comes. Both methods use data provided by Statistics Finland. They use micro level data on private enterprises in Finland during the years 1989 – 2016. The market share approach requires establishment level data and information on the revenue and employment. The indirect inference method uses the same data aggregated on firm level for the years 1993 – 2013. In addition, the simulation requires total factor productivity growth rate for the given years. The results suggest that 0.489 percentage points of growth has been missed yearly in Finland during the years 1989 – 2016 when calculated with revenue data. The missed growth was estimated to be 0.532 percentage points per year with employment data. The results are comparable in magnitude with the results from the United States and France. The magnitude has remained stable over the years. The indirect inference method suggests that most of the growth comes from incumbent own innovation: 59.3% in 1993 – 2003 and 57.8% in 2003 – 2013. The rest is due to creative destruction and new product varieties either by incumbents or entrants. If 0.5 percentage points of growth is missed every year, it would have had significant effects on the economy. For example, many social benefits are tied to price indices and over estimation of them would mean that the benefits have not risen as much as they should have. Given the systematic nature of the bias, the central bank should consider increasing its inflation target. The statistical offices that produce the price statistics may be able to lower the bias if they manage to keep up to date with incumbent own innovations, since the majority of growth is originating from it. Also chain linked index helps lowering the bias by updating the sample and weights on a yearly basis. Additional research is needed in order to find solutions to overcome the bias caused by creative destruction and imputation of missing prices.
-
(2018)Logistic regression has been the most common credit scoring model for several decades. The purpose of a credit scoring model is to distinguish good applicants from bad applicants so that the consumer credit can be lent to a person who is likely to repay it. In Finland, households' indebtedness has increased while wage development has stagnated. In addition to mortgage, indebtedness has increased because of the rising number of consumer credit loans. Consumer credit is usually unsecured loans, which are provided by several financial institutions quickly and flexible. Consumer credit is considered to be one of the major causes of default. Systematic risks are still being avoided for now, but the increased number of customers and the fierce competition in the sector can bring new risks that should be anticipated, as insolvent customers are making losses to financial institutions. Developing and deploying new credit scoring models is one of the best ways to hedge against default risks. The prediction accuracy and performance of tree-based credit scoring models have been studied. In many cases, tree-based algorithms have performed better than traditional statistical models such as the earlier mentioned logistic regression. In this master's thesis classical logistic regression is compared to these tree-based algorithms. The most well-known tree-based algorithms have been chosen, which are random forest, discrete Adaboost, real Adaboost, LogitBoost, Gentle Adaboost and Gradient Boosting. These methods use the tree algorithm as the base learner but differ in their iterative processes. The data that has been gathered from a Finnish medium-sized financial company, consists of customer's personal information and their payment behavior of sales finance. It is important to compare how different models predict insolvency in the light of different test statistics. In this thesis, the best-performing models are logistic regression and the Gradient Boosting algorithm. From my research's point of view, it is recommended to develop a credit scoring model based on the Gradient Boosting algorithm. This algorithm discloses different explanatory variables compared to logistic regression. These variables can explain better the causes of insolvency. The results are robust and plausible, because the different tests give similar conclusions.
-
(2018)Tässä tutkielmassa tutkitaan empiirisesti euroalueen optimaalisuutta valuutta-alueena aikaisempaan teoreettiseen kirjallisuuteen nojaten. Tarkastelussa keskitytään kriteeriin, jonka mukaan yhteisvaluutan muodostavien maiden tulisi olla suhdannesykleiltään yhteneväiset. Mundell (1961) määritteli useita kriteereitä optimaalista valuutta-aluetta muodostettaessa ja tässä tutkielmassa on valittu edellä mainittu yksi kriteeri tarkasteltavaksi. Tutkimuksessa hyödynnetään logaritmoitua asukaskohtaista bruttokansantuotetta mittaavaa neljännesvuosittaista aineistoa ja se on jaettu kahteen osaan. Ensimmäinen ajanjakso kattaa ajan ennen euroa (1960:1-1998:4) ja toinen ajanjakso kattaa ajan euron luomisen jälkeen (1999:1-2016:4). Aineisto kattaa euro12-maat, joihin kuuluvat Itävalta, Belgia, Suomi, Saksa, Ranska, Kreikka, Irlanti, Italia, Luxemburg, Alankomaat, Portugali ja Espanja. Tutkimusmenetelminä hyödynnetään yhteisintegroituvuusanalyysia. Maat on analysoitu pareittain niin, että Saksa on aina toisena. Saksa on valittu kontrollimaaksi, koska se on ollut yksi nopeimmin kasvavista euromaista euron luomisen jälkeen. Tutkimuksen perusteella Saksan ja muiden euromaiden väliltä ei löytynyt merkittävässä määrin yhteisintegroituvuussuhteita. Tämä indikoi, että tutkimuksessa käytettävien euromaiden suhdannesyklit eivät ole yhteneväiset Saksan kanssa, eikä euron luominen ole luonut konvergenssia euroalueelle. Keskeisimpänä johtopäätöksenä voidaan todeta, että tämän tutkimuksen perusteella euroalueen maat eivät ole suhdannesykleiltään yhteneväisiä, eikä euroalue siten täytä ainakaan toistaiseksi kaikkia optimaalisen valuutta-alueen kriteereitä. Tässä tiivistelmässä käytetyt lähteet: Mundell, R.A. 1961. A theory of optimum currency areas, American Economic Review, 657-665.
Now showing items 1-3 of 3