1. |
- Karlsson, Peter S.
(author)
-
Issues of incompleteness, outliers and asymptotics in high dimensional data
- 2011
-
Doctoral thesis (other academic/artistic)abstract
- This thesis consists of four individual essays and an introduction chapter. The essays are in the field of multivariate statistical analysis of High dimensional data. The first essay presents the issue of estimating the inverse covariance matrix alone and when it is used within the Mahalanobis distance in High-dimensional data. Three types of ridge-shrinkage estimators of the inverse covariance matrix are suggested and evaluated through Monte Carlo simulations. The second essay deals with incomplete observations in empirical applications of the Arbitrage Pricing Theory model and the interest is to model the underlying covariance structure among the variables by a few common factors. Two possible solutions to the problem are considered and acase study using the Swedish OMX data is conducted for demonstration. In the third essay the issue of outlier detection in High-dimensional data is treated. A number of point estimators of the Mahalanobis distance are suggested and their properties are evaluated. In the fourth and last essay the relation between the second central moment of a distribution to its first raw moment is considered in an financial context. Three possible estimators are considered and it is shown that they are consistent even when the dimension increases proportionally to the number of observations.
|
|
2. |
- Karlsson, Joel, 1980-
(author)
-
Issues of Complex Hierarchical Data and Multilevel Analysis : Applications in Empirical Economics
- 2012
-
Doctoral thesis (other academic/artistic)abstract
- This thesis consists of four individual essays and an introduction chapter. The essays are in the field of multilevel analysis of economic data. The first essay estimates capitalisation effects of farm attributes, with a particular focus on single farm payments (SFP), into the price of farms. Using a sample of Swedish farm transactions sold all across the country, the results from a spatial multiple-membership model suggests that the local effect of SFP is negative while there is a positive between-region effect of SFP, on farm prices. The second essay investigates the extent to which differences in the probability to exit from part-time unemployment to a full-time job can be accounted for by spatial contextual factors and individual characteristics. To correctly incorporate contextual effects, a multilevel analysis was applied to explore whether contextual factors account for differences in the probability of transition to full-time employment between individuals with different characteristics. The results indicate that there is a contextual effect and that there are some spatial spill-over effects from neighbouring municipalities. The third essay investigates the determinants of educational attainment for third-generation immigrants and natives in Sweden. Using a mixed-effects model that includes unobserved family heterogeneity, for linked register data, the main result is that the effect of parent’s educational attainment is mainly due to the between-parental education effect of family income. The fourth and last essay presents a new robust strategy for performance evaluation in the case of panel data that is based on routinely collected variables or indicators. The suggested strategy applies a cross-classified, mixed-effect model. The strategy is implemented in two illustrative empirical examples, and the robustness is investigated in a Monte Carlo study.
|
|
3. |
- Månsson, Kristofer
(author)
-
Issues of multicollinearity and conditional heteroscedasticy in time series econometrics
- 2012
-
Doctoral thesis (other academic/artistic)abstract
- This doctoral thesis consists of four chapters all related to the field of time series econometrics. The main contribution is firstly the development of robust methods when testing for Granger causality in the presence of generalized autoregressive conditional heteroscedasticity (GARCH) and causality-in-variance (i.e. spillover) effects. The second contribution is the development of different shrinkage estimators for count data models which may be used when the explanatory variables are highly inter-correlated.The first essay investigated the effect of spillover on some tests for causality in a Granger sense. As a remedy to the problem of over-rejection caused by the spillover effects White’s heteroscedasticity consistent covariance matrix is proposed. In the second essay the effect of GARCH errors on the statistical tests for Granger causality is investigated. Here some wavelet denoising methods are proposed and by means of Monte Carlo simulations it is shown that the size properties of the tests based on wavelet filtered data is better than the ones based on raw data.In the third and fourth essays ridge regression estimators for the Poisson and negative binomial (NB) regression models are investigated respectively. Then finally in the fifth essaya Liu type of estimator is proposed for the NB regression model. By using Monte Carlo simulations it is shown that the estimated MSE is lower for the ridge and Liu type of estimators than maximum likelihood (ML).
|
|
4. |
- Kraus, Katrin, 1979-
(author)
-
On the Measurement of Model Fit for Sparse Categorical Data
- 2012
-
Doctoral thesis (other academic/artistic)abstract
- This thesis consists of four papers that deal with several aspects of the measurement of model fit for categorical data. In all papers, special attention is paid to situations with sparse data.The first paper concerns the computational burden of calculating Pearson's goodness-of-fit statistic for situations where many response patterns have observed frequencies that equal zero. A simple solution is presented that allows for the computation of the total value of Pearson's goodness-of-fit statistic when the expected frequencies of response patterns with observed frequencies of zero are unknown.In the second paper, a new fit statistic is presented that is a modification of Pearson's statistic but that is not adversely affected by response patterns with very small expected frequencies. It is shown that the new statistic is asymptotically equivalent to Pearson's goodness-of-fit statistic and hence, asymptotically chi-square distributed.In the third paper, comprehensive simulation studies are conducted that compare seven asymptotically equivalent fit statistics, including the new statistic. Situations that are considered concern both multinomial sampling and factor analysis. Tests for the goodness-of-fit are conducted by means of the asymptotic and the bootstrap approach both under the null hypothesis and when there is a certain degree of misfit in the data. Results indicate that recommendations on the use of a fit statistic can be dependent on the investigated situation and on the purpose of the model test. Power varies substantially between the fit statistics and the cause of the misfit of the model. Findings indicate further that the new statistic proposed in this thesis shows rather stable results and compared to the other fit statistics, no disadvantageous characteristics of the fit statistic are found.Finally, in the fourth paper, the potential necessity of determining the goodness-of-fit by two sided model testing is adverted. A simulation study is conducted that investigates differences between the one sided and the two sided approach of model testing. Situations are identified for which two sided model testing has advantages over the one sided approach.
|
|