##### (Click Category to List Courses)

## 29 - DAB- Data Analytics & Business Intelligience

## DAB 301 - Proportion, Risk, Odds, Regression and Logistic Regression

Code | Start Date | Duration | Venue | |
---|---|---|---|---|

DAB 301 | 07 November 2022 | 5 Days | Istanbul | Registration Form Link |

DAB 301 | 05 December 2022 | 5 Days | Istanbul | Registration Form Link |

DAB 301 | 09 January 2023 | 5 Days | Istanbul | Registration Form Link |

DAB 301 | 06 March 2023 | 5 Days | Istanbul | Registration Form Link |

DAB 301 | 01 May 2023 | 5 Days | Istanbul | Registration Form Link |

DAB 301 | 12 June 2023 | 5 Days | Istanbul | Registration Form Link |

DAB 301 | 21 August 2023 | 5 Days | Istanbul | Registration Form Link |

DAB 301 | 16 October 2023 | 5 Days | Istanbul | Registration Form Link |

DAB 301 | 11 December 2023 | 5 Days | Istanbul | Registration Form Link |

Please contact us for fees

#### Course Description

The most desirable demand of the people worked in health area is to interpret the situations derived from

the any phases period of sickness and patients symptomatic complaints and expectations related with

recover their health in a short times. The matter likely is about results which will realize at the future or will be realized at the past by the studies whether are prospective or retroprospective.

Statistics presents some successful techniques which are about the problem. One of them is the regression method. In other words, it is the technique that predicts the best fitting values for the actual observations respectively. Regression technique can be modeled according to the problem parameters like expected outcomes and complained symptoms that interact with each other. They are assessed and converted to the regression model. The parameters are appreciated for the numerical or categorical result(s).

Next part of the course is about the subset of the regression analysis which is complementary support based on the situations which involve categorical observations and categorical predictions.

This course will explain to the trainee the regression analysis models are simple, multiple and logistic in order of difficulty. The presentation of the course will provide a simple, smart and summarize statistical explanations for the people employed in health and social disciplines. However the people in other disciplines can utilize the richness of the course.

#### Course Objectives

At the end of course, the trainee will be to:

- Analyse the situation statiscally in parameters,
- Predict the parameters expected,
- Interpret the outcomes of the situations,
- Understand how the parameters likely vary in confidence intervals,
- Design a 2x2 crosstable about the results of case control study surveyed,
- Learn how to extract the statistics subjected from 2x2crosstable,
- Interpret the results of the case control study surveyedm,
- And prepare a statistical report for case control study.

#### Who Should Attend?

- Physicians and Managers who are fighting to solve health and hospital problems involve statistical matters
- Statistical Fresh Researchers studying in case control studies(cohort studies)
- Anyone who wants to learn the statistical procedures to evaluate the comparison of nominal results associated with each other

#### Course Details/Schedule

#### Day 1

Regression Analysis (Prediction and Precision)

- Introduction
- The relationship between two data sets, Scatter diagrams
- The best fitting value of Y variable for any of the X given variable.
- The linearity of relationship between the variables X and Y.
- The pearson correlation, understanding of Regression,
- The variables are dependent and independent variable
- Regression Analysis, Find the regression equation and draw the best fitted line for the data.
- Examples*

#### Day 2

Multiple Regression Analysis, Introduction and definition.

- Multiple Regression Analysis , A multiple regression model
- The coefficients of multiple regression.
- The degrees of freedom, actual values ans predicted values,
- Residuals and minimizing the sum of squared errors, the least squares regression equation.
- Limitations of Multiple Regression
- The steps are used in multiple regression analysis
- The multicollinearity.
- Standard Deviation of Errors , Coefficient of Multiple Determination
- The coefficient of determination
- Factorial Models on Multiple Regression
- Models(Types) of Regression
- Polynomial regression
- Response Surface Regression
- Mixture Surface Regression
- Linear Regression Variable Selection Methods
- A stepwise, Backward Elimination, Forward Selection.
- Examples*

#### Day 3

Logistic Regression

- Introduction to Logistic Regression
- The Model, The Log-likelihood statistic (LL)
- Assessing Changes in Models
- Assessing Predictors: The Wald Statistic
- Similar to t-statistic in Regression.
- Assessing Predictors: The Odds Ratio(OR) and Exp(b)_in SPSS
- Types of Logistic Process
- Forced Entry, Stepwise, Hierarchical
- Assumptions from Linear Regression:
- Linearity, Independence of Errors, Multicollinearity
- Reporting the Analysis
- Multinomial Logistic Regression
- Examples*

#### Day 4

- Review of Descriptive Statistics ; variable, ratio, proportion, risk, relation
- Review of probability for binomial results
- Case studies ; Why and How are they organised ?
- Cross-tabulation for categorical data
- Resulting case studies ; Design of 2x2 cross table
- Success and Failure analysis through 2x2 cross tables
- Computing the statistics ; ratio, proportion, risk, relative risk, absolute relative risk
- To interpret the Risk is whether increased or decreased
- Odds and Odds Ratio
- To interpret the Odds Ratio for 2x2 cross table
- The assessment of statistics Relative Risk and Odds Ratio
- Examples*

#### Day 5

- The confidence interval for the Odds Ratio
- Reporting session of 2x2 cross tables proceed
- To use some softwares written on the concept of Odds Ratio on-line
- Coding 2x2 cross table of case control study as the data for SPSS or Minitab and entering data
- To evaluate the concept by the output of the softwares tried
- Practise by examples*