Regression Analysis by Example

By Terri-lyn Moore. Technometrics, 2001

2,080 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/f450ee71c0330d733ce01551bc1df3a88d66e87f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Statistics and Probability

Citations in the corpus

A global relationship between the heterotrophic and autotrophic components of soil respiration? (2004, methodology): “…one study (Thierron & Laudelout, 1996) with extremely high reported values for RS and RC was excluded after model fitting, as an outlier test (Chatterjee & Price, 1991) using it and the reduced data set was highly significant (Y*5 817, YPRED5 191, SE(YPRED)5 105, T535 5.96,…”
Investor reaction to women directors (2010, background): “5 is not considerably larger than 1 (Chatterjee et al., 2000).”
Context-aware recommender systems (2008, neutral): “, using numerous existing feature selection procedures from machine learning [42], data mining [48], and statistics [28], based on existing ratings data during the data preprocessing phase.”
COVID-19 information overload and generation Z's social media discontinuance intention during the pandemic lockdown (2021, background): “The variance inflation factor for each independent variable is lower than the suggested threshold of 4, thus suggesting the absence of multicollinearity (Chatterjee and Hadi, 2015; Harman, 1967; Podasakoff et al., 2003).”
Some New Diagnostics of Multicollinearity in Linear Regression Model (2019, result): “Correct detection by these new indicators also followed the results from the existing literature (Chatterjee & Hadi 2006; Gujarati & Porter 2008; Maddala 1988).”
Revisiting Burns And Stalker: Formal Structure And New Venture Performance In Emerging Economic Sectors (2006, methodology): “Because role formalization, functional specialization, founding team size, and the interaction term between founding team size and functional specialization were highly correlated, the variance inflation factor for our regressions exceeded the threshold for multicollinearity recommended by Chatterj”
Comparing floral nectar and aphid honeydew diets on the longevity and nutrient levels of a parasitoid wasp (2004, background): “The slopes of the regression lines for females given nectar and honeydew were significantly different (t = 3.95, d.f. = 202, P 0.0001, t = difference in slopes/standard error of regression coefficient estimates, Chatterjee & Price, 1977).”
Pattern-based local linear regression models for short-term load forecasting (2016, methodology): “The multiple linear regression (MLR [20]) model for STLF is of the form:”
The Other Pathway to the Boardroom: Interpersonal Influence Behavior as a Substitute for Elite Credentials and Majority Status in Obtaining Board Appointments (2006, background): “The highest variance inflation factor (VIF) was less than ten, and the mean VIF was not significantly greater than one in all models, suggesting that multicollinearity was not a problem (Chatterjee, Hadi, and Price, 2000).”
The Confounding Effect of Class Size on the Validity of Object-Oriented Metrics (2001, methodology): “This means that tstatistics for significance testing tend to be small and confidence intervals tend to be wide [28][89].”
Quantile Correlations and Quantile Autoregressive Modeling (2012, positive): “In the classical regression model, one can construct the partial correlation to measure the linear relationship between variables Y and X after adjusting for (or controlling for) vector Z (e.g., see Chatterjee and Hadi 2006).”
Can models of presence‐absence be used to scale abundance? Two case studies considering extremes in life history (2005, neutral): “We assumed that multcollinearity was influential when VIF /10 or the mean of all VIFs were considerably /1 (Chatterjee et al. 2000).”
Changing places: A cross-language perspective on frequency and family size in Dutch and Hebrew (2005, methodology): “We compare the more complex model with two predictors with the simpler model with only one predictor (the family size ratio) using a likelihood ratio test for the by-participant analysis (Crawley, 2002, p. 696) and the standard F test for the by-item analysis ( Chatterjee, Hadi, & Price, 2000, pp. 7”
Microarray data normalization and transformation (2002, neutral): “Recall that logarithms treat numbers and their reciprocals symmetrically: log2(1) = 0, log2(2) = 1, log2( (1)⁄2) = −1, log2(4) = 2, log2( (1)⁄4) = −2, and so on.”