Statistical Analysis and Modeling with SPSS

Learn basic correlation theory, interpret positive, negative, and zero correlations, and implement SPSS to compute Pearson coefficients, scatter plots, and covariance using mutual funds data.

Analyze correlation patterns among sectoral funds and other classes, highlighting diversification from negatively correlated funds like itec and ibf, and noting higher risk in large-cap, mid-cap, and elss funds.

Learn to use the SPSS data editor to generate simple and matrix scatter plots, assess strong positive and weak negative correlations, and justify trends with visuals.

Explore how to create and interpret a simple scatter plot in SPSS, recognize near-zero and positive correlation patterns, and annotate visuals with text boxes.

Analyze the correlation between heart pulse before run and after run for smokers and non-smokers, yielding a strong positive correlation (0.603) and a SPSS scatter plot.

Explore how to create scatter plots in the statistics viewer, add a regression and reference lines, and interpret positive, negative, and zero correlations along with R square values.

Explore how height relates to heart pulse before and after a run using SPSS, generating scatter plots and correlation coefficients to reveal weak negative relationships and fit lines.

Analyze correlations in heart pulse before and after a run for smokers and non-smokers, and the relation between height and heart pulse, using scatter plots and dataset conclusions.

Assess the correlation between work events and health, using Pearson’s r 0.429, a scatter plot, and a linear fit to show how work hassles relate to depression.

Explore using SPSS to assess the correlation between hassles and support, visualize it with a scatter plot, and interpret a weak negative Pearson correlation around -0.31 linked to depression.

Explore correlation analysis in SPSS using scatter plots and a bivariate Pearson correlation to show weak positive relationships in small data sets with fit lines.

Explore linear regression in SPSS, from the y = mx + c theory to dependent and independent variables, slope, intercept, and r square, with p-values, t-values, and stock returns.

Perform linear regression in SPSS. Use reliance returns as the dependent variable and Sensex returns as the independent variable, yielding y = mx + c with m = 1.16.

Explore the regression of stock return on the BSE Sensex, interpreting r-square, t-values, p-values, and the equation r_il = 1.116 × r_bse.

Explore how to interpret t value, p value, and R square in a regression of Reliance returns on Sensex data, with case A insights, positive coefficients, and correlation.

Create scatter plots with a linear fit in SPSS to derive and interpret the regression equation and R-squared for Sensex returns against reliance and Infosys.

Create a template and list attributes for the independent variable. Interpret regression outputs, including t values, p values, and R-squared, to relate Sensex returns to Infosys returns.

create a scatter plot of infosys versus sensex, add a fit line, note a positive, flat slope with r-squared of 0.098, and observe high data density indicating low volatility.

This lecture demonstrates linear regression in SPSS to predict heart rate from body temperature, delivering the equation heart rate = 4.398 × body temperature − 88.101 and noting statistical significance.

Prepare SPSS data by adjusting variable properties and views, then generate and interpret a Pearson correlation matrix for mutual funds' net asset values and test two-tailed significance at 0.01.

Explore simple linear regression in SPSS with a copper expansion case, modeling expansion as 0.021 times temperature plus 7.449, where temperature explains about 68.9% of the variation.

Analyze copper expansion using regression to show that each kelvin rise in temperature increases expansion by about 7.47 units, with a strong 0.83 correlation and 68.9% of variation explained.

Create a scatter plot of copper expansion versus temperature (Kelvin) to show a correlation (r ≈ 0.83) and a regression line with intercept 7.449 and slope 0.021; report R-squared 0.689.

Explore how machine settings influence energy consumption using SPSS, identify dependent and independent variables, apply stepwise linear regression, derive the regression equation, and note a negative correlation with low R-squared.

Examine how the regression links machine settings to energy consumption, noting that a one-unit rise reduces energy by 0.317 units while the intercept remains significant.

Create a scatter plot of energy consumption versus machine settings in SPSS, add fit line and negative correlation, and interpret the regression equation with a positive y-intercept and low R-squared.

Assess debt as a function of income using linear regression in SPSS with a debt to income dataset of 31 customers, noting r-squared 0.038 and an insignificant income coefficient.

Explore regression analysis in SPSS to assess debt against income. Identify an insignificant negative relationship with low R and R^2, and a scatter plot showing a weak fit.

Learn how existing credit card debt predicts debt to income ratio with a two-variable linear regression in SPSS, showing a 0.647 correlation and an R-squared of 0.419.

Explore how credit card debt raises the debt-to-income ratio through simple linear regression, with coefficients, p-values, r-squared of 41.9%, and scatter plots illustrating strong positive correlation.

Demonstrate predicting debt-to-income ratios from credit card debt using a regression equation in Excel, with output = 1.814 times debt + 7.282, showing input changes drive predictions.

Learn how to generate predicted values from a regression equation using Excel, input random independent values, and visualize results with a scatter plot, with SPSS as an alternative.

Explore multiple regression by interpreting r squared, coefficients, t values, slopes, and the intercept, assess predictor impact on the dependent variable, and address multicollinearity with SPSS data and MySQL outputs.

Explore the key output variables in multiple regression, including r square, p values for X1..Xn, and collinearity diagnostics, illustrated with cement data and heat evolved from components X1-X4, using SPSS.

Using SPSS, this lecture demonstrates multiple regression on cement data to study how aluminates, silicates, trisilicate, and ferrite affect heat evolved.

Explore multiple regression results, interpreting r-square, coefficients, and p-values to identify illuminate as a significant predictor of heat evolved, while ferrite, silicate, and trisilicate show weaker or negative relationships.

Analyze a multiple regression equation to interpret how aluminate and silicate affect heat evolved. Ferrite reduces heat evolved; trisilicate shows negative correlations, and aluminate is the main significant predictor.

Walk through a multiple regression example using cement chemistry data to model heat evolved, detailing data import, descriptive statistics, the regression equation, and predicting values for different ingredient inputs.

Learn to build and interpret a multiple regression in SPSS, predicting before-run and after-run heart pulse from height and weight in smokers.

In this case, smokers’ heart pulse is poorly predicted by height or weight. With an R-squared of 14.4% and insignificant t-values, the model offers a weak fit for predictive modeling.

Enter and compare the before run and after run regression equations for heart pulse with height and weight, and conclude the smokers model has low r-square and insignificant t-values.

Conduct a two-variable multiple regression in SPSS to show how existing credit card debt and other debt influence debt-to-income ratio, including model fit, coefficients, and significance.

Conduct a multiple regression of debt-to-income ratio on existing credit card debt and other debt, interpreting slopes (0.656, 1.510) and R-squared around 55% with a scatter plot and trend line.

Explore a predictive regression that shows existing credit card debt and other debt increase the debt-to-income ratio by 0.656 and 1.510 per unit, with 55.3% variance and all terms significant.

Explore multiple regression in SPSS, focusing on debt to income ratio, scatterplots, predicted values, and model diagnostics (t-values, p-values, r-squared) with credit card debt and other debt.

In SPSS, perform a multiple regression with extroversion as the dependent variable and age and weekly car miles as independent variables, yielding a regression equation and an R-squared of 51.2%.

In this multiple regression, extroversion increases by 0.442 with age and 0.582 with car miles, with R² = 51.2% and both predictors significant.

Analyze how age and car miles predict extroversion using SPSS regression and descriptive statistics. Note that car miles carry a larger coefficient than age, impacting predicted values.

Explore logistic regression concepts, focusing on binary and dichotomous independent variables, model equations, interpretation of significance, multicollinearity, and SPSS implementation with sample data.

Explore using the IBM SPSS statistics data editor to build a logistic regression model, focusing on Wald statistics, and defining independent and dependent variables with dichotomous outcomes.

Use the SPSS data editor to run binary logistic regression predicting loan default from income, debt, and education, with education as a reference category, reviewing the output and model fit.

Formulates and displays a regression model in IBM SPSS Viewer, specifying dependent and independent variables such as household income, debts, and education levels, and derives the regression equation and coefficients.

Explore regression analysis in SPSS, identify dependent and independent variables, and interpret significance through p-values and Wald statistics, with 0.05 as a common threshold.

Use Excel to compute the existing debt-to-income and income-to-debt ratios, identify high-income, low-debt customers as low default risk, and frame a logistic regression approach to credit card approval.

Analyze how height, weight, and running activity influence after run heart pulse using SPSS logistic regression, considering gender as a binary variable in a heart pulse study.

Explore binary logistic regression in SPSS with heart pulse data, modeling gender as the dependent variable using height and weight as covariates, and interpreting Wald tests and Hosmer-Lemeshow test.

Explore how to restructure regression equations in SPSS to predict heart pulse after running, deriving gender-specific models using height, weight, and running as predictors.

Explore how height and weight influence after-run pulse in smokers using regression equations, assess significance, and identify a best-fit model across male and female smokers.

Describe how SPSS constructs a smoking gender equation to predict heart pulse from height and weight, using descriptive statistics and regression models.

Apply binary logistic regression in SPSS to model gold deposits using arsenic and lead as predictors and other metals as a dummy variable, interpreting Hosmer-Lemeshow and classification plots.

Explore how to use a regression model to predict gold deposits from arsenic and lead, using Wald significance, correlation insights, and predictive value calculations in SPSS.

Interpret the regression results: arsenic and lead positively relate to gold deposits, while non-gold metals negatively affect gold; Wald coefficients and predicted values assess the model.

Learn how multinomial logistic regression extends logistic regression to non-dichotomous outcomes in SPSS, using factors and covariates to model categorical activity levels with numeric predictors like height and weight.

Import a data set in SPSS and set up multinomial logistic regression. Use heights and weights as covariates and smoking and gender as categorical variables, selecting the reference category.

Learn how to build a multinomial logistic SPSS model with two factors and two covariates, including intercept, using likelihood ratio tests and classification tables for model fit.

Explore data quality via the case processing summary and warnings, then interpret multinomial regression outputs using chi-square, likelihood ratio tests, and Wald indicators for model fit and variable selection.

Evaluate data quality and model fitting information for regression analysis; note high warnings and 65% zero-frequency cells undermine fit, with insignificant independent variables per Wald test results.

Analyze the asymptotic correlation matrix in SPSS to assess a multinomial logistic regression model, showing negative or weak correlations and overall insignificance of independent variables for best fit.

This lecture explains applying multinomial logistic regression to predict educational course choices—general, academic, or vocational—from socioeconomic status and reading ability as a covariate.

This SPSS lecture analyzes parameter estimates to identify y intercept, reading and writing scores, middle income, and socioeconomic status as predictors for general and vocational regression, yielding multinomial logistic models.

Explore how regression equations and parameter estimates in SPSS model general and vocational courses, using reading scores, writing scores, and socioeconomic status to assess significance.

Explore asymptotic correlation matrix and metrics in SPSS, focusing on parameter estimates, significance, and regression equations to interpret general and vocational course results.

Interpret the output to assess model fit, key tests, and the significance of reading and writing scores for general and vocational courses in SPSS.

Interpret parameter estimates for general and vocational course models, identifying significant predictors (y-intercept, reading and writing scores, middle income, socioeconomic status) and confirming best-fit regression and multinomial logistic regression outputs.

Analyze multinomial logistic regression estimates for general and vocational programs, showing how one-unit changes in reading, writing, and socioeconomic status alter relative log odds against the academic reference.

Interpret multinomial logistic regression results by analyzing relative log odds between academic, general, and vocational programs, and assess how writing and reading scores influence these comparisons.