Determining Innocence in Innocent-Spouse Court Cases Using Logit/Probit Analysis
In: Advances in Taxation, Band 17
2236 Ergebnisse
Sortierung:
In: Advances in Taxation, Band 17
SSRN
In: Kölner Zeitschrift für Soziologie und Sozialpsychologie: KZfSS, Band 64, Heft 2, S. 377-395
ISSN: 1861-891X
In: Kölner Zeitschrift für Soziologie und Sozialpsychologie: KZfSS, Band 64, Heft 2, S. 377-395
ISSN: 0023-2653
Logit- und Probitregression werden als multivariate Analyseverfahren zur Analyse von dichotomen abhängigen Variablen in den Sozialwissenschaften routinemäßig eingesetzt. Beide Verfahren können so interpretiert werden, dass sich aus einer linearen Modellierung einer unbeobachteten Variabley* eine nichtlineare Modellierung der Wahrscheinlichkeiten füry = 1 ergibt. Wir zeigen erstens, dass diese Nichtlinearität im Vergleich zu linearen Regressionsverfahren zu Problemen bei der Interpretation der Modellergebnisse führt. Insbesondere die in der logistischen Regression häufig verwendeten odds ratios (exponierte Logit-Koeffizienten) sind unseres Erachtens problematisch. Stattdessen empfehlen wir neben graphischen Interpretationshilfen die Verwendung von (korrigierten) durchschnittlich marginalen Effekten (AME). Zweitens zeigen wir anhand einer Serie von Monte-Carlo-Simulationen, dass die üblichen Regressionskoeffizienten bei Logit- und Probitanalysen nicht zwischen verschachtelten Modellen verglichen werden können. Da in den Sozialwissenschaften bei der Modellbildung jedoch häufig schrittweise vorgegangen wird, wäre ein Verfahren, das einen validen Vergleich von Effektstärken zwischen den Modellen erlaubt, sehr nützlich. Wie wir anhand unserer Simulationsstudie zeigen, führen durchschnittlich marginale Effekte und Koeffizienten, die nach dem Vorschlag von Karlson et al. (Sociological Methodology 42, 2012) korrigiert wurden, in sehr verschiedenen Situationen zu gültigen Ergebnissen.y*-standardisierte Koeffizienten sind für einen Modellvergleich hingegen weniger geeignet und Koeffizienten eines linearen Wahrscheinlichkeitsmodells sollten ausschließlich bei normalverteilten Variablen verwendet werden.
In: Quantitative applications in the social sciences 45
In: Sage university papers
In: Sage University papers
In: Quantitative applications in the social sciences 138
In: IZA Discussion Paper No. 10530
SSRN
In: American journal of political science, Band 38, Heft 1, S. 230
ISSN: 1540-5907
In: American journal of political science: AJPS, Band 38, Heft 1, S. 230-255
ISSN: 0092-5853
In: Social science research: a quarterly journal of social science methodology and quantitative research, Band 109, S. 102802
ISSN: 1096-0317
In: Working paper - Institute for Policy Analysis, University of Toronto no. 7706
For a large variety of discrete choice models (or contingency table models) efficientand stable maximum likelihood methods can be constructed basedon the majorization method. The course introduces majorization methods for algorithm construction. We show how to use the majorization principle to reduce complicated optimization problems to sequences of weighted or unweighted least squares problems. Majorization methods are then applied to data analysis techniques used in economics, political science, psychometrics, ecology, sociology, and education.
BASE
In: American journal of political science, Band 36, Heft 3, S. 762
ISSN: 1540-5907
In: American journal of political science: AJPS, Band 36, Heft 3, S. 762-784
ISSN: 0092-5853
Though some regression analysts find the R2 statistic relatively useless, others use it extensively to evaluate model performance. In probit & logit analyses, the lack of an analog to the ordinary least squares (OLS) R2 statistic is problematic, & several pseudo-R2s have been proposed to help in the evaluation of model performance. Dichotomizing a continuous interval-level variable results in distortions due to a loss of information, the degree to which these distortions affect the pseudo-R2s vis-a-vis the OLS R2, which is based on the underlying continous dependent variable, is unknown. Here, simulation techniques are employed to compare four common pseudo-R2s for probit & logit with the R2 that would be obtained under OLS regression. After making a correction to one of the measures, two are found to compare quite favorably with the OLS R2. It is concluded that the choice between them may be simply a matter of availability & ease of use. 4 Tables, 2 Figures, 22 References. Adapted from the source document.
Abstract. The globalization movements that started towards the end of the 20th century influenced many areas of the economy. As a result of the globalization, despite the persistence of the political borders, countries have established borderless relationships in the economic arena. On the other hand, along with globalization, financial crises have become more frequent in the world economy. In particular, the fact that the 2008 Global Crisis reached serious dimensions made it necessary to take measures to stabilize the markets and to evaluate the factors that would shake the market and make the market fragile. Financial fragility is a concept that is often concurrently used with the concepts of financial instability and financial crisis. Financial fragility is a hypothesis which was developed by Hyman Minsky and is different from both concepts and also is interactively affected by instability and crises. In this study, unlike other studies with reference to Minsky's financial fragility hypothesis, we aimed to identify the factors that make fragile the financial markets in Turkey. Logit and probit models were studied with the data of 1990:01-2018:05 period. With reference to the study results, the increase in ratios of the volume of bank loans and M2 occur reserves increases the possibility of future crises. Besides, it is found at the end of the study that a decrease in the composite leading indicators index and in M2 in the BIST 100 index will strengthen the probability of a crisis.Keywords. Financial fragility, Minsky hypothesis, Logit model, Probit model.JEL. G00, C12, B23.
BASE
In: Social work research & abstracts, Band 27, Heft 3, S. 16-21