Advanced Statistical Methods and Optimization

Author
Affiliation

Prof. Dr. Tim Weber

Deggendorf Institute of Technology

Preface

This is the script for the lecture “Advanced Statistical Methods and Optimization” at the DIT/Campus Cham. I do realize, that this body of knowledge has been repeated over and over, but have decided to do my own nonetheless so I can add my own flavor to the realms of statistics. This work is heavily inspired by (Wickham and Grolemund 2016). Please note that this material is copyrighted, you are not allowed to copy, at least ask for permission - you are likely to get it.

Tim Weber, Oct. 2024

Glossary

Text Abbreviations

ANOVA

Analysis of Variance

CI

Confidence Interval

CL

Confidence Level

CDF

cumulative function

CLT

Central Limit Theorem

CTQ

Critical To Quality

dof

degree of freedom

DoE

Design of Experiments

EDA

Exploratory Data Analysis

FN

false negative

FP

false positive

gof

goodness of fit

H0

Null Hypothesis

Ha

Alternative Hypothesis

IQR

Interquartile Range

KPI

Key Performance Indicator

KS

Kolmogorov Smirnov

LLN

Law of Large Numbers

MLE

Maximum Likelihood Estimation

MSA1

Measurement System Analysis Type I

UCL

Upper Control Limit

LCL

Lower Control Limit

UWL

Upper Warning Limit

LWL

Lower Warning Limit

PCC

Pearson Correlation Coefficient

PDF

Probability Density Function

PMF

Probability Mass Function

PoI

Parameter of Interest

p

Population proportion

ppm

parts per million

QC

Quality Control

QQ

Quantile-Quantie

SE

Standard Error

TTF

Time to failure

TN

true negative

TP

true positive

.w.r.t

with respect to

Z

Z-standardization

Symbol Abbreviations

\(\alpha\)

significance level

\(\beta\)

false negative risk

\(\epsilon\)

residuals

\(\mu_0\)

the true mean of a population

\(\varphi(x)\)

probability density function

\(\phi(x)\)

cumulative probability density function or cumulative distribution function

\(\sigma_0^2\)

the true variance of a population

\(\sigma_0\)

the true standard deviation of a population

\(C_g\)

potential Measurement System Capability Index

\(C_{gk}\)

Measurement Capability Index with systematic error

\(C_p\)

potential process capability

\(C_{pk}\)

actual process capability including centering

\(k\)

number of predictors in a model

\(MSE\)

mean squared errors

\(n\)

number of data points/observations in the sample

\(N\)

number of datapoints/observations in the population

\(P\)

Probabilities

\(r^2\)

Coefficient of determination

\(r^2_{adjusted}\)

adjusted Coefficient of determination

\(sd\)

the standard deviation of a dataset

\(SSE\)

Sum of squared errors as calculated by

\(x_i\)

the individual datapoints

\(\bar{x}\)

the mean value of the datas

\(X\)

Predictor Variable

\(Y\)

Response Variable

\(\hat{y}\)

predicted value

\(y_i\)

true value