Learn about a comprehensive framework of the right skills that you can master to be a successful Data Analyst
What you will learn
course aims to provide and enhance predictive modelling skills across business sectors
The course picks theoretical and practical datasets for predictive analysis
Observations, interpretations, predictions and conclusions are explained then and there on the examples as we proceed through the training
The course also emphasizes on the higher order regression models such as quadratic and polynomial regressions
Description
Logistic regression in SPSS is defined as the binary classification problem in the field of statistic measuring. The difference between a dependent and independent variable with the guide of logistic function by estimating the different occurrence of the probabilities, i.e., it is used to predict the outcome of the independent variable (1 or 0 either yes/no) as it is an extension of a linear regression which is used to predict the continuous output variables.
Logistic regression is a technique used in the field of statistics measuring the difference between a dependent and independent variable with the guide of logistic function by estimating the different occurrence of probabilities. They can be either binomial (has yes or No outcome) or multinomial (Fair vs poor very poor). The probability values lie between 0 and 1, and the variable should be positive (<1).
It targets the dependent variable and has the following steps to follow:
- n- no. of fixed trials on a taken dataset.
- With two outcomes trial.
- The outcome of the probability should be independent of each other.
- The probability of success and failures must be the same at each trial.
Predictive modelling course aims to provide and enhance predictive modelling skills across business sectors/domains. Quantitative methods and predictive modelling concepts could be extensively used in understanding the current customer behavior, financial markets movements, and studying tests and effects in medicine and in pharma sectors after drugs are administered. The course picks theoretical and practical datasets for predictive analysis. Implementations are done using SPSS software. Observations, interpretations, predictions and conclusions are explained then and there on the examples as we proceed through the training. The course also emphasizes on the higher order regression models such as quadratic and polynomial regressions which arenβt covered in other online courses
ο Essential skillsets β Prior knowledge of Quantitative methods and MS Office, Paint
ο Desired skillsets — Understanding of Data Analysis and VBA toolpack in MS Excel will be useful
The course works across multiple software packages such as SPSS, MS Office, PDF writers, and Paint.
Regression modelling forms the core of Predictive modelling course. The core objective of this course is to provide skills in understand the regression model and interpreting it for predictions. The associated parameters of the regression model will be interpreted and tested for significance and test the goodness of fit of the given regression model.
Through this course we are going to understand:
- Interpretation of regression attributes such as R-Squared (correlation coefficient), t and p values
- m (slope) and c (intercept),
- Dependent variables (Y), independent (A1, A2, A3β¦β¦) variables, and Binary/Dummy B1, B2, B3 β¦..) variables
- Examining significance/relevance of A, B variables for regression model (equation) goodness of fit
- Predicting Y-variable upon varying values of A, B variables
- Understanding Multi-Collinearity and its disadvantages
- Implementation on sample datasets using SPSS and output simulation in MS Excel
Content