Testing for Serial Correlation in Regression Models with Lagged Dependent Variables

1993 ◽  
Vol 75 (4) ◽  
pp. 716 ◽  
Author(s):  
Robert K. Rayner
Keyword(s):  
Regression Models ◽  
Serial Correlation ◽  
Dependent Variables

scholarly journals The Influence of Industrial Clusters and Place on Innovation and Entrepreneurial Behaviour

2007 ◽  
Vol 8 (3) ◽  
pp. 165-174 ◽  
Author(s):  
Pam McRae-Williams ◽  
Julian Lowe ◽  
Peter Taylor
Keyword(s):  
Regression Models ◽  
Personal Characteristics ◽  
Industrial Clusters ◽  
Factor Scores ◽  
Standard View ◽  
Behaviour Factor ◽  
Regional Clusters ◽  
Entrepreneurial Behaviour ◽  
Dependent Variables ◽  
Central Result

Responses from a questionnaire survey of wine and tourism businesses operating in regional clusters were analysed using factor analysis. These suggested three factor scores relating to entrepreneurial behaviour; four factor scores relating to cluster activities and attributes; and three factors relating to the respondents' personal characteristics. The three entrepreneurial behaviour factor scores were interpreted as: innovator, calculator and venturer. These were used as dependent variables in regression models. The independent variables were the cluster and personal characteristics factor scores, industry and place. The central result was that the cluster activity variables did not have a significant impact on the innovator behaviour variable, which contradicts the standard view. Cluster activities and attributes were found to attract entrepreneurs of the calculator kind, and to a lesser extent, of the venturer kind. Place did seem to offer an attraction to entrepreneurs beyond those offered by the intensities of the cluster activities and attributes.

Women in the Legislatures and Executives of the World: Knocking at the Highest Glass Ceiling

1999 ◽  
Vol 51 (4) ◽  
pp. 547-572 ◽  
Author(s):  
Andrew Reynolds
Keyword(s):  
Multivariate Regression ◽  
Regression Models ◽  
Socioeconomic Development ◽  
Party System ◽  
Political Representation ◽  
Parliamentary System ◽  
Dependent Variables ◽  
The World ◽  
Political Office ◽  
Women In Parliament

This article reports the results of a survey of women in legislatures and executives around the world as they were constituted in 1998 (N = 180). The chief hypotheses regarding the factors hindering or facilitating women's access to political representation were tested by multivariate regression models. The regression models juxtaposed a cocktail of institutional, political, cultural, and socioeconomic variables with the following dependent variables: (1) the percentage of MPs who are women and (2) the percentage of cabinet ministers who are women.A number, although not all, of the cited hypotheses were statistically confirmed and more finely quantified. The socioeconomic development of women in society has an effect on the number of women in parliament but not in the cabinet. A country's length of experience with multipartyism and women's enfranchisement correlates with both the legislative and the executive percentage. Certain electoral systems are more women friendly than others. The ideological nature of the party system affects the number of women elected and chosen for cabinet posts. And last, the state's dominant religion, taken as a proxy for culture, also statistically relates to the number of women who will make it to high political office. However, other long-held hypotheses were not proved. The degree of democracy is not a good indicator of the percentage of women who will make it into the legislature or the cabinet, nor is the dichotomy between a presidential or parliamentary system.

scholarly journals A Regionalised Neyman-Scott Model of Rainfall with Convective and Stratiform Cells

1997 ◽  
Vol 1 (1) ◽  
pp. 71-80 ◽  
Author(s):  
P. S. P. Cowpertwait ◽  
P. E. O'Connell
Keyword(s):  
Cluster Model ◽  
Regression Models ◽  
Parameter Estimates ◽  
Model Parameter ◽  
Lack Of Fit ◽  
Dependent Variables ◽  
The Uk ◽  
Poisson Cluster ◽  
Residual Errors ◽  
Urban Sites

Abstract. A single-site Neyman-Scott Poisson cluster model of rainfall, with convective and stratiform cells, is fitted to data for 112 sites scattered throughout the UK using harmonic variables to account for seasonality. The model is regionalised by regressing the estimates of the harmonic variables on site dependent variables (e.g. altitude) to enable rainfall to be simulated at any ungauged site in the UK. An assessment of the residual errors indicates that the regression models can be used with reasonable confidence for urban sites. Furthermore, the regional variations of the model parameter estimates are found to be in agreement with meteorological knowledge and observation. Simulated I h extreme rainfalls are found to compare favourably with observed historical values, although some lack-of-fit is evident for higher aggregation levels.

Relationships Between GIS Environmental Features and Adolescent Male Physical Activity: GIS Coding Differences

2006 ◽  
Vol 3 (2) ◽  
pp. 230-242 ◽  
Author(s):  
Russell Jago ◽  
Tom Baranowski ◽  
Michael Harris
Keyword(s):  
Physical Activity ◽  
Environmental Variables ◽  
Regression Models ◽  
Adolescent Male ◽  
Grocery Stores ◽  
Activity Levels ◽  
Independent Variables ◽  
Environmental Features ◽  
Gis Data ◽  
Dependent Variables

Background:It is not clear if relationships between GIS obtained environmental features and physical activity differ according to the method used to code GIS data.Methods:Physical activity levels of 210 Boy Scouts were measured by accelerometer. Numbers of parks, trails, gymnasia, bus stops, grocery stores, and restaurants within the commonly used 400 m and 1-mile (1609.3 m) buffers of subject residences and distance to the nearest feature were calculated. Residential density, connectivity, and crime rate were calculated. Regression models with minutes of sedentary, light, or moderate-to-vigorous activity as dependent variables and environmental and demographics as independent variables were run with backward deletion of environmental variables.Results:Park, crime, and gym variables were associated with physical activity, but relationships varied according to whether a 400 m, 1 mile, or nearest criteria was used.Conclusion:Environmental variables were associated with the physical activity of adolescent males, but the association was method dependent.

Do college athletics marketers convert social media growth into ticket sales?

2017 ◽  
Vol 18 (2) ◽  
pp. 212-227 ◽  
Author(s):  
Nels Popp ◽  
Chad McEvoy ◽  
Nicholas Watanabe
Keyword(s):  
Social Media ◽  
Regression Models ◽  
College Athletics ◽  
Percentage Increase ◽  
Sport Organizations ◽  
Single Measure ◽  
Content Type ◽  
Dependent Variables ◽  
Social Media Engagement ◽  
The Impact

Purpose The purpose of this paper is to examine the relationship between growth in social media engagement, as defined by annual percentage increase in Facebook Likes and Twitter Followers, of US college athletics departments and outcome metrics of attendance and ticket revenue. Design/methodology/approach Regression models were developed to determine the amount of variance in dependent variables (attendance and ticket revenue) could be explained by several independent variables, including team success, team history, conference affiliation, Facebook Likes, and Twitter Followers. Four years of data were collected for each variable. Findings The regression models predicted between 53 and 88 percent of the variance among dependent variables. Social media measures, however, were not statistically significant predictors of attendance or ticket revenue. Research limitations/implications The number of Facebook Likes and Twitter Followers were used as a proxy measure of social media engagement. While growth in Likes and Followers are a popular and convenient gauge of social media engagement, they represent a single measure of a multi-faceted construct. Also, data were limited to public university athletics departments, which are required to disclose annual ticket revenue. Findings may not be generalizable to other sport organizations. Practical implications The findings suggest growing social media interactions may not necessarily achieve marketing objectives related to increasing attendance or ticket revenue. Originality/value While numerous studies have examined the impact of social media on sport organizations, no prior studies have attempted to draw empirical connections between social media marketing efforts and revenue measures within sport organizations. This study represents the first to begin to examine this relationship.

scholarly journals Timeliness of Service Delivery for Children With Later-Identified Mild-to-Severe Hearing Loss

2014 ◽  
Vol 23 (1) ◽  
pp. 116-128 ◽  
Author(s):  
Elizabeth A. Walker ◽  
Lenore Holte ◽  
Meredith Spratford ◽  
Jacob Oleson ◽  
Anne Welhaven ◽  
...  
Keyword(s):  
Hearing Loss ◽  
Early Intervention ◽  
Service Delivery ◽  
Regression Models ◽  
Neonatal Period ◽  
Intervention Programs ◽  
Newborn Period ◽  
Dependent Variables ◽  
Newborn Hearing

Purpose In this study, the authors examined diagnostic and intervention services for children identified with hearing loss (HL) after the newborn period. Method The authors compared ages at service delivery and length of delays between service delivery steps for 57 later-identified children with HL and 193 children who referred for assessment from the newborn hearing screen (NHS). For only later-identified children, regression models were used to investigate relationships among predictor variables and dependent variables related to service delivery. Results Children who referred from the NHS received follow-up services at younger ages than later-identified children. Later-identified children had significantly longer delays from HL confirmation to entry into early intervention, compared to children who referred from the NHS. For later-identified children, degree of HL predicted ages at follow-up clinical services. Children with more severe HL received services at younger ages compared to children with milder HL. Gender predicted the length of the delay from confirmation to entry into early intervention, with girls demonstrating shorter delays. Conclusions The current results lend support to the need for ongoing hearing monitoring programs after the neonatal period, particularly when children enter early intervention programs because of language/developmental delays.

Sign in / Sign up

Export Citation Format

Share Document