## Introduction to Spatial-Temporal Statistics

April 27, 2017

9:00am - 5:00pm

DBH 4011

Instructors: Greg Britten, Yara Mohajerani

## Introduction

Data collected in time and/or space exhibit unique properties that require attention to draw proper conclusions from statistical analyses. In this workshop, students are introduced to statistical concepts that are particularly useful for analyzing spatial-temporal data.

Who: This course is targeted primarily at graduate students and researchers who are interested in applied spatial-temporal data analysis

Requirements: Participants must bring a laptop with a few specific software packages installed (see Pre-Workshop Instructions).

Prerequisites: Previous experience with R or Python and knowledge of basic statistical concepts (at least one statistics course or experience statistically analyzing data).

## Tentative Schedule

Time
8:30-9:00 Sign-in (coffee & bagels)
Come early to fix pre-install issues
9:00-9:30 Jupyter Notebooks, R/Python packages, GitHub repo
9:30-11:00 Introduction to Spatial-Temporal Statistics
11:00 - 11:15 Break (coffee)
11:15-12:30 Applied Time Series Analysis
12:30-1:00 Lunch
1:00-2:30 Spatial Analysis and Model Selection
2:30-2:45 Break (coffee)
2:45-4:30 Spatial Interpolation and Regression

## Syllabus

1. Introduction/Preparation
• Jupyter notebooks
• R packages: ‘gstat’, ‘sp’; Python interface via package ‘rpy2’
2. Introduction to Spatial-Temporal Statistics
• Autocorrelation function and uncertainty quantification
• Periodicity (seasonality, daily cycles, etc.)
3. Applied Time Series Analysis
• Trend analysis
• Harmonic regression
• Time series decomposition
• Model selection
4. Applied Spatial Analysis
• Spatial autocorrelation, variogram functions
• Spatial interpolation (Kriging)
• Spatial regression
• Model selection

## Pre-Workshop Instructions

See the course’s GitHub page for instructions: https://github.com/UCIDataScienceInitiative/SpaceTime We’ll expect you to have the Anaconda Python distribution installed with version 2.7 activated: https://www.continuum.io/downloads