The Bayesian estimation of private investment in Finland
Abstract
This paper estimates an investment equation for private investment using Bayesian estimation techniques. In the paper we derive the optimal capital accumulation behavior in the model economy from the households’ optimization problem of utility. The equation is derived as in Smets and Wouters (2003). The model contains costly adjustment of investment and random shocks to adjustment cost function. The driving variable of investment is Tobin Q variable.
The empirical proxy for Tobin Q in this paper is the ratio of OMX Helsinki Cap Index to the price index of the physical capital. The investment series is the seasonally adjusted private investment in quarterly national accounts.
The AR(1) modelled investment shocks are found to be less persistent in Finland than in the euro area. The estimated median of persistence parameter for Finland is 0.485. Also the shocks to investment adjustment cost function are found to vary less in Finland as in the euro area. The estimated standard deviation of the shocks is 0.065. The adjustment cost parameter is roughly the same for both data sets. The results are robust to loosening the strict prior of discount factor,
beta=0.99. The paper also provides discussion about adjustment cost parameter and we investigate the behaviour of the posterior chain of B with different prior distributions for the parameter.
Tiivistelmä
Tässä pro gradussa estimoidaan yhtälö yksityisille investoinneille bayesilaisella menetelmällä. Tässä työssä optimaalinen pääoman akkumulointi mallikansantaloudessa johdetaan kotitalouksien hyödyn optimointi-ongelmasta. Investointiyhtälö johdetaan kuten Smets’n ja Wouterin (2003) artikkelissa. Malli sisältää investointien sopeutuskustannukset ja satunnaisia shokkeja sopeutuskustannus-funktioon. Investointien selittävä muuttuja on Tobin Q -muuttuja.
Empiirinen vastine teoreettiselle Tobin Q muuttujalle on OMX Helsinki Cap indexin arvo suhteutettuna fyysisen pääoman hintaindeksillä. Työssä käytetty investointisarja on kausitasoitettu yksityisten investointien sarja kansantalouden neljännestilinpidossa.
Investointishokit ovat AR(1)-prosessi. Shokit osoittautuvat vähemmän pysyviksi Suomessa kuin euroalueella. Estimoitu AR(1)-kerroin investointishokeille on 0.485. Investointishokit myös vaihtelevat vähemmän Suomessa kuin euroalueella, sillä estimoitu shokkien keskihajonta on 0.065. Investointien sopeutuskustannus on likipitäen samankokoinen Suomessa ja euroalueella. Tulokset ovat robusteja kiinnitetyn diskonttausparametrin beta=0.99 löysäämiselle antamalla betalle eri
priorijakaumia. Tässä työssä myös keskustellaan sopeutuskustannusparametrista ja tutkitaan sen posterioiriketjujen käyttäytymistä kun sille annetaan eri priorijakaumia.
...
Keywords
Metadata
Show full item recordCollections
- Pro gradu -tutkielmat [29052]
Related items
Showing items with similar title or keywords.
-
How is social impact bonds model used by public sectors in impact investing in Finland?
Chen, Jianhong (2022)This is a paper to look into how Social Impact Bonds model is used by public sectors in impact investing in Finland. Finland has made a number of impact investments to promote the nation’s well-being in different themes ... -
A Bayesian Reconstruction of a Historical Population in Finland, 1647–1850
Voutilainen, Miikka; Helske, Jouni; Högmander, Harri (Springer, 2020)This article provides a novel method for estimating historical population development. We review the previous literature on historical population time-series estimates and propose a general outline to address the well-known ... -
“The Chinese Will Not Change; We Have To Change” : Adjustment of the Finns to the Chinese in a Chinese Investment Facilitation Context
Stopniece, Santa (Immigrant Institutet, 2017)This paper explores the intercultural communication, cultural adjustment strategies and power relations between Finnish and Chinese representatives in the context of Chinese investment facilitation in Finland. The study ... -
Price Optimization Combining Conjoint Data and Purchase History : A Causal Modeling Approach
Valkonen, Lauri; Tikka, Santtu; Helske, Jouni; Karvanen, Juha (University of Pennsylvania Press, 2024)Pricing decisions of companies require an understanding of the causal effect of a price change on the demand. When real-life pricing experiments are infeasible, data-driven decision-making must be based on alternative data ... -
Bayesian applications in dynamic econometric models
Luoto, Jani (University of Jyväskylä, 2009)The purpose of this thesis is to provide a few new ideas to the field of Bayesian econometrics. In particular, the focus of the thesis is on analyzing dynamic econometric models. In the first essay, we provide an easily ...