We investigate the importance of ethnic origin and local labour markets conditions for self-employment propensities in Sweden. In line with previous research, we find differences in the self-employment rate between different immigrant groups as well as between different immigrant cohorts. We use a multilevel regression approach in order to quantify the role of ethnic background, point of time for immigration and local market conditions in order to further understand differences in self-employment rates between different ethnic groups. We arrive at the following: The self-employment decision is to a major extent guided by factors unobservable in register data. Such factors might be, that is, individual entrepreneurial ability and access to financial capital. The individual’s ethnic background and point of time for immigration play a smaller role for the self-employment decision but are more important than local labour market conditions.
In this paper, we suggest a unit root test for a system of equations using a spectral variance decomposition method based on the Maximal Overlap Discrete Wavelet Transform. We obtain the limiting distribution of the test statistic and study its small sample properties using Monte Carlo simulations. We find that, for multiple time series of small lengths, the wavelet-based method is robust to size distortions in the presence of cross-sectional dependence. The wavelet-based test is also more powerful than the Cross-sectionally Augmented Im et al. unit root test (Pesaran, M. H. 2007. "A Simple Panel Unit Root Test in the Presence of Cross-section Dependence." Journal of Applied Econometrics 22 (2): 265-312.) for time series with between 20 and 100 observations, using systems of 5 and 10 equations. We demonstrate the usefulness of the test through an application on evaluating the Purchasing Power Parity theory for the Group of 7 countries and find support for the theory, whereas the test by Pesaran (Pesaran, M. H. 2007. "A Simple Panel Unit Root Test in the Presence of Cross-section Dependence." Journal of Applied Econometrics 22 (2): 265-312.) finds no such support. © 2019 Walter de Gruyter GmbH, Berlin/Boston.
In this paper, we suggest a unit root test for a system of equations using a spectral variance decomposition method based on the Maximal Overlap Discrete Wavelet Transform. We obtain the limiting distribution of the test statistic and study its small sample properties using Monte Carlo simulations. We find that, for multiple time series of small lengths, the wavelet-based method is robust to size distortions in the presence of cross-sectional dependence. The wavelet-based test is also more powerful than the Cross-sectionally Augmented Im et al. unit root test (Pesaran, M. H. 2007. "A Simple Panel Unit Root Test in the Presence of Cross-section Dependence." Journal of Applied Econometrics 22 (2): 265-312.) for time series with between 20 and 100 observations, using systems of 5 and 10 equations. We demonstrate the usefulness of the test through an application on evaluating the Purchasing Power Parity theory for the Group of 7 countries and find support for the theory, whereas the test by Pesaran (Pesaran, M. H. 2007. "A Simple Panel Unit Root Test in the Presence of Cross-section Dependence." Journal of Applied Econometrics 22 (2): 265-312.) finds no such support.
The multiple time series and ridge regression techniques are proposed for modeling and analyzing a scaled real life (or a simulated) data as a SUR model with VAR(p) disturbances. The regression coefficients are estimated via the generalized least squares method if collinearity is weak and otherwise the regression coefficients are estimated by the generalized ridge regression method. Small sample likelihood ratio test statistic and model selection criteria are employed for selecting the smallest possible lag order for the VAR process. Moreover, Monte Carlo simulations (1000 replications) are conducted to examine the properties of some new and some of the existing ridge parameters in rectifying the collinearity problem in SUR models with VAR(2) disturbances via the trace(MSE) and condition number criteria. Two data sets are analyzed to illustrate the findings of the article.
This paper utilizes Wavelet based methodology to estimate and test for trends and granger causality in temperature andprecipitation. We use quarterly data from Sweden for the period 1884 up to 2011. The analysis suggests that temperatureand precipitation in Sweden currently have a positive trend in 2011. Thus the recent lower levels of the variables 2009-2010are estimated to be temporary fluctuations or deviations from the trend. Moreover, in the short run there are feedbackeffects between the variables and over longer periods, 4-8 years, temperature granger cause precipitation.
In this paper we outline a framework for forecasting using maximal overlap discrete wavelet transform (MODWT) based multiresoulution analysis (MRA). This framework has been applied for forecasting the tourism arrival series from Denmark to Norway. We compare forecasted values obtained from modeling the data in the time domain with the forecasted values from the wavelet domain using the traditional Box-Jenkins methodology. In both cases, diagnostic tests have been conducted to insure the specification of the model. The results have shown that the wavelet based forecasts outperforms the traditional Box-Jenkins approach in term of forecasts accuracy.
This article introduces two different non-parametric wavelet-based panel unit-root tests in the presence of unknown structural breaks and cross-sectional dependencies in the data. These tests are compared with a previously suggested non-parametric wavelet test, the parameteric Im-Pesaran and Shin (IPS) testand a Wald type of test. The results from the Monte Carlo simulations clearly show that the new wavelet-ratio tests are superior to the traditional tests both interms of size and power in panel unit-root tests because of its robustness to cross-section dependency and structural breaks. Based on an empirical Central American panel application, we can, in contrast to previous research (where bias due to structural breaks is simply disregarded), find strong, clear-cut support for purchasing power parity (PPP) in this developing region.
This article introduces two different non-parametric wavelet-based panel unit-root tests in the presence of unknown structural breaks and cross-sectional dependencies in the data. These tests are compared with a previously suggested non-parametric wavelet test, the parameteric Im-Pesaran and Shin (IPS) test and a Wald type of test. The results from the Monte Carlo simulations clearly show that the new wavelet-ratio tests are superior to the traditional tests both in terms of size and power in panel unit-root tests because of its robustness to cross-section dependency and structural breaks. Based on an empirical Central American panel application, we can, in contrast to previous research (where bias due to structural breaks is simply disregarded), find strong, clear-cut support for purchasing power parity (PPP) in this developing region.
We investigate the importance of ethnic origin and local labour markets conditions for self-employment propensities in Sweden. In line with previous research we find differences in the self-employment rate between different immigrant groups as well as between different immigrant cohorts. We use a multilevel regression approach in order to quantify the role of ethnic background, point of time for immigration and local market conditions in order to further understand differences in self-employment rates between different ethnic groups. We arrive at the following: The self-employment decision is to a major extent guided by factors unobservable in register data. Such factors might be i.e. individual entrepreneurial ability and access to financial capital. The individual’s ethnic background and point of time for immigration play a smaller role for the self-employment decision but are more important than local labour market conditions.
Part-time work is one of the most well-known « atypical » working time arrangements. In contrast to previous studies focusing on the supply side, the originality of our research is to investigate the demand-side of part-time work and to examine how and why companies use part-time work. Based on a large and unique sample of European firms operating in 21 member states, we use a multilevel multinomial modeling in a Bayesian environment. Our results suggest that the variations in the extent of part-time workers at the establishment level is determined more by country-specific features than by industry specific factors.
This paper presents a Seemingly Unrelated Regressions estimation of earnings differentials between three generations of immigrants and natives in Sweden. The results show that male first-generation immigrants were at an earnings advantage compared to male natives. Among male second-generation immigrants the earnings differentials compared to natives were very small, while third-generation immigrants were at an earnings disadvantage compared to natives. The same pattern was found among females. Thus, the results indicate that ethnic differences in earnings are likely to occur even after several generations spent in a country and that the problem of immigrant assimilation that exists in many European countries may last for several generations.
This article tests the home-country self-employment hypothesis on immigrants in Sweden. The results show that the self-employment rates vary between different immigrant groups but we find no support for the home-country self-employment hypothesis using traditional estimation methods. However, when applying quantile regression method we find such evidence when testing results from the 90th quantile. This indicates that home-country self-employment traditions are important for the self-employment decision among immigrant groups with high self-employment rates in Sweden. Furthermore, the result underlines the importance of utilizing robust estimation methods when the home-country self-employment hypothesis is tested.
In this paper we generalize four tests of multivariate linear hypothesis to panel data unit root testing. The test statistics are invariant to certain linear transformations of data and therefore simulated critical values may conveniently be used. It is demonstrated that all four tests remains well behaved in cases of where there are heterogeneous alternatives and cross-correlations between marginal variables. A Monte Carlo simulation is included to compare and contrast the tests with two well-established ones.
In this paper, a short background of the Jarque and McKenzie (JM) test for non-normality is given, and the small sample properties of the test is examined in view of robustness, size and power. The investigation has been performed using Monte Carlo simulations where factors like, e.g., the number of equations, nominal sizes, degrees of freedom, have been varied.
Generally, the JM test has shown to have good power properties. The estimated size due to the asymptotic distribution is not very encouraging though. The slow rate of convergence to its asymptotic distribution suggests that empirical critical values should be used in small samples.
In addition, the experiment shows that the properties of the JM test may be disastrous when the disturbances are autocorrelated. Moreover, the simulations show that the distribution of the regressors may also have a substantial impact on the test, and that homogenised OLS residuals should be used when testing for non-normality in small samples.
Experimental studies often measure an individual’s quality of life before and after an intervention, with the data organized into a square table and analyzed using matched pair modeling. However, it is not unusual to find missing data in either round (i.e., before and/or after) of such studies and the use of multiple imputations with matched-pair modeling remains relatively unreported in the applied statistics literature. In this paper we introduce an approach which maintains dependency of responses over time and makes a match between the imputer and the analyst. We use ‘before’ and ‘after’ quality-of-life data from a randomized controlled trial to demonstrate how multiple imputation and matched-pair modeling can be congenially combined, avoiding a possible mismatch of imputation and analyses, and to derive a properly consolidated analysis of the quality-of-life data. We illustrate this strategy with a real-life example of one item from a quality-of-life study that evaluates the effectiveness of patients’ self-management of anticoagulation versus standard care as part of a randomized controlled trial.
Two Level (TL) models allow the total variation in the outcome to be decomposed as level one and level two or ‘individual and group’ variance components. Two Level Mixture (TLM) models can be used to explore unobserved heterogeneity that represents different qualitative relationships in the outcome.
In this paper, we extend the standard TL model by introducing constraints to guide the TLM algorithm towards a more appropriate data partitioning. Our constraints-based methods combine the mixing proportions estimated by parametric Expectation Maximization (EM) of the outcome and the random component from the TL model. This forms new two level mixing conditional (TLMc) approach by means of prior information. The new framework advantages are: 1. avoiding trial and error tactic used by TLM for choosing the best BIC (Bayesian Information Criterion), 2. permitting meaningful parameter estimates for distinct classes in the coefficient space and finally 3. allowing smaller residual variances. We show the benefit of our method using overweight and obesity from Body Mass Index (BMI) for students in year 6. We apply these methods on hierarchical BMI data to estimate student multiple deprivation and school Club effects.
In this article, we propose a general framework for performance evaluation of organizations and individuals over time using routinely collected performance variables or indicators. Such variables or indicators are often correlated over time, with missing observations, and often come from heavy-tailed distributions shaped by outliers. Two new double robust and model-free strategies are used for evaluation (ranking) of sampling units. Strategy 1 can handle missing data using residual maximum likelihood (RML) at stage two, while strategy two handles missing data at stage one. Strategy 2 has the advantage that overcomes the problem of multicollinearity. Strategy one requires independent indicators for the construction of the distances, where strategy two does not. Two different domain examples are used to illustrate the application of the two strategies. Example one considers performance monitoring of gynecologists and example two considers the performance of industrial firms.
This paper applies wavelet multi-resolution analysis (MRA), combined with two types of causality tests, to investigate causal relationships between three variables: real oil price, real interest rate, and unemployment in Norway. Impulse response functions were also utilised to examine effects of innovation in one variable on the other variables. We found that causal relations between the variables tend to be stronger as the wavelet time scale increases; specifically, there were no causal relationships between the variables at the lowest time scales of one to three months. A causal relationship between unemployment rate and interest rate was observed during the period of two quarters to two years, during which time a feedback mechanism was also detected between unemployment and interest rate. Causal relationships between oil price and both interest rate and unemployment were observed at the longest time scale of eight quarters. In conjunction with Granger causality analysis, impulse response functions showed that unemployment rates in Norway respond negatively to oil price shocks around two years after the shocks occur. As an oil exporting country, increases (or decreases) in oil prices reduce (or increase) unemployment in Norway under a time horizon of about two years; previous studies focused on oil importing economies have generally found the inverse to be true. Unlike most studies in this field, we decomposed the implicit aggregation for all time scales by applying MRA with a focus on the Norwegian economy. Thus, one main contribution of this paper is that we unveil and systematically distinguish the nature of the time-scale dependent relationship between real oil price, real interest rate, and unemployment using wavelet decomposition.
Model selection criteria are often used to find a "proper" model for the data under investigation when building models in cases in which the dependent or explained variables are assumed to be functions of several independent or explanatory variables. For this purpose, researchers have suggested using a large number of such criteria. These criteria have been shown to act differently, under the same or different conditions, when trying to select the "correct" number of explanatory variables to be included in a given model; this, unfortunately, leads to severe problems and confusion for researchers. In this paper, using Monte Carlo methods, we investigate the properties of four of the most common criteria under a number of realistic situations. These criteria are the adjusted coefficient of determination (R2-adj), Akaike's information criterion (AIC), the Hannan–Quinn information criterion (HQC) and the Bayesian information criterion (BIC). The results from this investigation indicate that the HQC outperforms the BIC, the AIC and the R2-adj under specific circumstances. None of them perform satisfactorily, however, when the degree of multicollinearity is high, the sample sizes are small or when the fit of the model is poor (i.e., there is a low R2) . In the presence of all these factors, the criteria perform very badly and are not very useful. In these cases, the criteria are often not able to select the true model.
This article analyzes the effects of multicollienarity on the maximum likelihood (ML) estimator for the Tobit regression model. Furthermore, a ridge regression (RR) estimator is proposed since the mean squared error (MSE) of ML becomes inflated when the regressors are collinear. To investigate the performance of the traditional ML and the RR approaches we use Monte Carlo simulations where the MSE is used as performance criteria. The simulated results indicate that the RR approach should always be preferred to the ML estimation method.