Calculus I and II Algebra Statistics

The course has two parts: Forecasting and reliability. In the first part you learn to forecast variables. For example you can forecast the evolution of a company's sales, or monthly unemployment in Spain. We will use univariate ARIMA models. In the second part you will learn to estimate the duration of processes and / or components. This is the basis of reliability analysis. We use parametric and nonparametric estimators for complete or censored data. The course is absolutely applied and each week there is a theoretical lecture and practical class in computer room where you learn to use the techniques learned.

1. Time Series Analysis. 1.1 Introduction. Characteristics of a time series: Trend, homoscedasticity and seasonal cycle. 1.2 Stationary Time Series. 1.3 Transformation on non Stationary Time Series into Stationary Time Series. 1.4 Simple and partial autocorrelation function. 1.5 Models AR (1) AR (2) and AR (p) 1.6 Models MA (1), MA (2) and MA (q) 1.7 ARMA Models 1.8 ARIMA Models 1.9 Estimation and diagnosis. 1.10 Forecasting 1.11 Seasonal ARIMA Models 12.1 Forecasting with seasonal ARIMA models 2. Reliability 2.1 Introduction to duration data (ADS) 2.1 Functions used: reliability function and failure rate 2.3 Types of failure rates. 2.4 Parametric models: Weibull 2.5 Graphical Methods to determinate the model. 2.6 Duration estimation for complete data. 2.7 Censored Data. Types of censorship. 2.8 Graphical methods for censored data. Kaplan Meier Estimator 2.9 Parametric Estimation with censored data. 2.10 Accelerated tests (under stress) 11.2 Series and parallel systems. Introduction to complex systems.

Every week there are two lessons. A theoretical one that introduces various analysis techniques, and a practical one (in a computer room) where the learned technique are applied to real problems.