GEOG5927 Predictive Analytics : Introduction

# GEOG5927 Predictive Analytics : Introduction
### Roger Beecham
### 28 Feb 2022

---

???

Looked at module website
Also on Minerva
1. Introduce Module
2. Say how it will run
3. Say a bit on timings
4. Look at the first bit of subtantive work -- practical 1
---

## Module content and philosophy

* Spatial modelling to simulate and predict consumer behaviour
  + Exploratory analysis
  + Microsimulation
  + Agent-based modelling

<br>

* Research case studies to evaluate modelling techniques in practice
  + Practical sessions
  + Individual data science report
  + Group data science presentation

???
* Introduces some key theory and techniques in modern data analysis.
* Has both technical and applied aspects.
* Techniques in which Geog dept specialises
* Doing so with datasets and domains relevant to marketing science

---

## Outcomes

By the end of this module you should be able to:

1. **Explain** and **critically evaluate** the role of spatial analytics in simulating and predicting consumer behaviours

2. **Apply** geocomputational modelling and simulation techniques on real data sets

3. **Devise** and **employ** spatial modelling tools to address business problems, presenting and justifying recommendations in an appropriate context

---

???

1 : theory
2 : application
3 : both

---
## Module team

<table>
 <thead>
  <tr>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> img </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> name </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> role </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> sessions </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> assignments </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  <html><body><img src="img/profile_roger.jpg" width="80" height="80"></body></html>
</td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Roger Beecham </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Convenor </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 1,2 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  <html><body><img src="img/profile_rachel.jpeg" width="80" height="80"></body></html>
</td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Rachel Oldroyd </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Lecturer </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 3 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> NA </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  <html><body><img src="img/profile_jg.jpg" width="80" height="80"></body></html>
</td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Jiaqi Ge </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Lecturer </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 4,5 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 2 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  <html><body><img src="img/profile_nik.jpg" width="80" height="80"></body></html>
</td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Nick Malleson </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Lecturer </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 4,5 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> NA </td>
  </tr>
</tbody>
</table>

---
## Scheduling

<table>
 <thead>
  <tr>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> session </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> wc </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> lecture </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> lab </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> recap </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 1 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 28 Feb </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> wed 1100-1800 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> fri 1700-1800 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 2 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 7 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> wed 1100-1800 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> fri 1700-1800 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 3 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 14 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> wed 1100-1800 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> fri 1700-1800 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 4 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 21 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> wed 1100-1800 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> fri 1700-1800 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 5 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 25 Apr </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> NA - pres </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> NA </td>
  </tr>
</tbody>
</table>
.small-font[
Additional drop-in sessions via Teams: <br>
-- Tuesdays 1:00pm–2:00pm and led by [Will James](https://environment.leeds.ac.uk/geography/staff/1054/dr-will-james).
]
???

Breakdown of activity : short and fat

---

## Outline : Lectures

<table>
 <thead>
  <tr>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> session </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> wc </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> academic </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> lecture </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 1 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 28 Feb </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> RB </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Simulating behaviour </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 2 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 7 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> RB </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Targeted marketing </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 3 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 14 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> RO </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Machine learning </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 4 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 21 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> JG/NM </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Agent-based modelling </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 5 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 25 Apr </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> JG/NM </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Close </td>
  </tr>
</tbody>
</table>
<br>
.small-font[
RB: Roger Beecham, RO: Rachel Oldroyd, JG: Jiaqi Ge, NM: Nick Malleson
]

.small-font[
Additional drop-in sessions via Teams: <br>
-- Tuesdays 1:00pm–2:00pm and led by [Will James](https://environment.leeds.ac.uk/geography/staff/1054/dr-will-james).
]
---

## Assessment

* Assignment 1: 75%
  + Individual data analysis based on labs 1, 2 (and 3)
  + 2,000 words, 4 figures
  + Thu 24th Mar 2022 by 2pm

<br>

* Assignment 2: 25%
  + Group presentations based on lab 4
  + Presentations for times session Wed 27th Apr 2022

---

## Assessment

<table>
 <thead>
  <tr>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> session </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> wc </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> academic </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> lecture </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> deadline </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 1 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 28 Feb </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> RB </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Simulating behaviour </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 2 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 7 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> RB </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Targeted marketing </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 3 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 14 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> RO </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Machine learning </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 4 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 21 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> JG/NM </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Agent-based models </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Ass 1 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 5 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 25 Apr </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> JG/NM </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Close </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Ass 2 </td>
  </tr>
</tbody>
</table>
<br>
.small-font[
RB: Roger Beecham, RO: Rachel Oldroyd, JG: Jiaqi Ge, NM: Nick Malleson
]

.small-font[
Additional drop-in sessions via Teams: <br>
-- Tuesdays 1:00pm–2:00pm and led by [Will James](https://environment.leeds.ac.uk/geography/staff/1054/dr-will-james).
]

---

## Assessment

<table>
 <thead>
  <tr>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> session </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> wc </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> academic </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> lecture </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> deadline </th>
   <th style="text-align:left;color: #616161 !important;background-color: #ffffff !important;font-size: 18px;"> progress </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 1 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 28 Feb </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> RB </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Simulating behaviour </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> #1 data </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 2 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 7 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> RB </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Targeted marketing </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> #1 analysis </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 3 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 14 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> RO </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Machine learning </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;">  </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> #1 analysis </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 4 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 21 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> JG/NM </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Agent-based models </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Ass #1 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> #2 material </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 5 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 25 Apr </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> JG/NM </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> Close </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> Ass #2 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> #2 surgery </td>
  </tr>
</tbody>
</table>
<br>
.small-font[
RB: Roger Beecham, RO: Rachel Oldroyd, JG: Jiaqi Ge, NM: Nick Malleson
]

.small-font[
Additional drop-in sessions via Teams: <br>
-- Tuesdays 1:00pm–2:00pm and led by [Will James](https://environment.leeds.ac.uk/geography/staff/1054/dr-will-james).
]

---

## How

---

## R Resources

.small-font[
* H. Wickham and G. Grolemund, R for Data Science, O’Reilly Media, 2017.
  + *The primer for doing data analysis with R.*
  + [Free online](https://r4ds.had.co.nz).

* R. Lovelace et al., Geocomputation with R, CRC Press, 2019.
  + *Comprehensively introduces spatial data handling in R.*
  + [Free online](https://geocompr.robinlovelace.net).

* K. Healy, Data Visualization: A Practical Introduction, Princeton University Press, 2018.
  + *Integrates ggplot2 code with key Information Visualization theory and using real social science datasets.*
  + [Free online -- draft version](https://socviz.co).
]
---

## Module Resources

* [Module website](https://www.roger-beecham.com/predictive-analytics/)
<br><br>
* [Module handbook](https://minerva.leeds.ac.uk/ultra/courses/_533807_1/cl/outline)
<br><br>
* [Assignment briefs](https://minerva.leeds.ac.uk/ultra/courses/_533807_1/cl/outline)
<br><br>
* [Assignment submission page](https://minerva.leeds.ac.uk/ultra/courses/_533807_1/cl/outline)

???

If not already, spend time on Module Handbook.

---

## Module Website

<!--

---

## How to Learn
--

* *Come to "remote" lectures and labs* <br>
--

* *Engage: Try stuff out, ask questions* <br>
--

* *Contribute to the module [Slack](https://predictive-analytics.slack.com)* <br>
--

* *Coursework throughout: lectures and labs*

---

# Introduction to Predictive Analytics

???

* That's introduction to module...
* Push on with a bit of theory…
* Why Predictive Analytics

---

## Data-driven science

**"big data"** and **"data science"** on Google Trends, Oct 2019

???

Know already, but from 2010s -- big data is a big industry -- transforming the way businesses, societies run
Calling upon new ways of doing science

---

## Data-driven science

* 100s years ago -- **theoretical science**
	+ Newton’s laws, Maxwell’s Equations

* <50 years ago -- **computational science**
	+ Simulate complex phenomena

* Today -- **data-intensive science**
  + Generate knowledge through 	observation
]

???

Tony Hey :

* Experimental science :  all we could know about was what we observed directly
* Theory : abstracting above observations to generate fundamental laws about how things are
* Sumulation : Test theories by simulating real world
* Today : Back observation, but deriving new ways in new ways

---

## Data-driven science

.small-font[
> *The next generation of scientific discovery will be data-driven as previously unrecognised patterns are discovered by analysing massive and mixed datasets.*
>
> David Willets MP, 2013, Then Minister for Universities and Science
]
--

---

## Data-driven science : example

???

Generate knowledge in new ways

* Real-time google searches tell us something about what is happening in the world.
* Association between how many people search for flu-related topics and how many people actually have flu symptoms.
* Can we use these to to estimate how much flu is circulating and predict outbreak?

---

## Data-driven science : example

[`Lazer et al. 2014`](https://science.sciencemag.org/content/343/6176/1203.full)

???

* Biases need to be more thoroughly thought through

---

## Data-driven science : example

???

* Generate new types of knowledge — newly opened up for analysis

*How* people choose to decorate their homes — is is culturally and geographically distinctive. Or in a globalised world does a typical home London decorate the same as a typical home in Tokyo?

Computer Vision  algorithms to automatically detect presence of certain ornaments and stylistic objects —see if there is spatial consistency in these…

---

## Data-driven science : example

.small-font[
[`fivethirtyeight.com/parole-assessment-simulator/`](https://projects.fivethirtyeight.com/parole-assessment-simulator/)
]

???

* Sentencing based typically on a criminal’s past crimes…
* But here on *likely* crimes that individuals might commit in the future.
* Based on statistical associations in past offender dataset
* Using statistical probabilities based on factors such as age, employment history and prior criminal record
* *Targeted* intervensions

---

## Data-driven science : this module

<br>
.pull-left[.small-font[
**Data mining and machine learning** — <br>
Detect hidden patterns in data

**Information Visualization** — <br>
Explore complex structure and patterns in data

**Predictive analytics** — <br>
Predict, under uncertainty, what will happen in future
]
]

---

## Data-driven science : assignments

<br>
.pull-left[.small-font[
.small-font[
**Assignment 1 -- ** <br> Generate a large synthetic dataset of customers <br>
and look for behavioural and demographic associations between individuals <br>
to better *target* marketing activity.

<br>

**Assignment 2 -- ** <br> Use data and heuristics to explore and predict <br>
how customers behave and respond to different store formats.
]]
]

---
class: center middle

# Session 1 : Simulating Behaviour

---

## Simulating behaviour practical

---

## Spatial microsimulation

.pull-left[.right[
`Survey data`
.small-font[
individual-level and rich in detail <br>
small sample and may be biased
]]
]

.pull-right[.left[
`Census data`
.small-font[
high-level and low in detail <br>
population-level and complete
]]
]

???

Many situations where interested in knowing population that lives in an area: interests, preferences, spending patterns

Are population-level datasets that exist.:

Census is amazing for counting people according to high-level characteristics...

But we only have a limited set of attribute information.

Lots missed off from it – commercial setting – interests and preferences.

Instead -- rely on comparatively small sample survey data for studying interests and preferences.

Spatial microsimulation allows us to match rich individual-level data to a population we know less about.

---

## Spatial microsimulation

<br><br><br>
.small-font[
> *The creation, analysis and modelling of individual-level data allocated to geographic zones.*
>
> Lovelace & Dumont 2016
]

---

## Spatial microsimulation

???

1. Start: survey of individuals and a geographic area. Each one of these spatial units is a small area of Leeds.
2. Allocate individuals from the survey to the small spatial units of Leeds. Lots of copies of individuals – such that they sum to the entire population in that area.

---

## Spatial microsimulation

???

Random allocation sounds not very sensible.

In microsim do so using some prior information on those spatial units – that you know about from a real dataset -- from the census in this case.

If Census says that 60% are orange, 30% black and 10% blue, copy individuals from the survey based data in those same proportions...

Notice arrows are varying in thickness to communicate this.

Census acts as constraints – give us a better approximation of the populations than just randomly allocating.

---

## Spatial microsimulation

---

## Spatial microsimulation

???

Blue: 40%
Black: 40%
Orange: 20%

---

## Spatial microsimulation

<br>
.pull-left[.right[
Microsimulation does not<br>
generate **new data** <br>
`-----`
]]

<br><br>
.pull-right[.left[
`-----` <br>
But **copies of existing data**
]]
---

## Spatial microsimulation: Examples

Health : Smoking <br>
  .small-font[[Tomintz et al. 2008](https://rgs-ibg.onlinelibrary.wiley.com/doi/abs/10.1111/j.1475-4762.2008.00837.x)]

--
.small-font[
 .pull-left[.right[
    **Why?** <br>
    Reported in individual surveys, but not population-level and not from place-to-place
    ]]]
--

.small-font[.pull-right[.left[
    **Benefits** <br>
    Could be used to target/locate smoking support clinics
  ]]]
--
  .small-font[.pull-right[.left[
    **"Benefits"** <br>
    Could be used by a Tobacco company for targeting investment
  ]]
  ]

???

Estimate levels of smoking in a population.

---

## Spatial microsimulation: Examples

Economics : Policy Evaluation <br>
  .small-font[[De Agostini et al. 2016](https://www.econstor.eu/bitstream/10419/197592/1/868840475.pdf)]

--
.small-font[
 .pull-left[.right[
    **Why?** <br>
    Simulate / spread impacts inferred from individual-level data over an entire country
    ]]]

--
  .small-font[
  .pull-right[.left[
    **Benefits** <br>
    Quantify (under uncertainty) the impacts of a regressive welfare reform at the country-level

]]]

--
  .small-font[.pull-right[.left[
    **Benefits** <br>
    Evidence-based decision-making
  ]]]

---

## Spatial microsimulation: Examples

Transport : Simulating travel behaviour <br>
  .small-font[[Lovelace et al. 2014](https://www.sciencedirect.com/science/article/pii/S0966692313001361)]

--
.small-font[
 .pull-left[.right[
    **Why?** <br>
    When designing infrastructure, want to know about the distribution of individuals meeting a particular set of characteristics

]]]
--
.small-font[
  .pull-right[.left[
    **Benefits** <br>
    provide evidence around likely winners and losers of a new infrastructure investment

]]
  ]
---

## Spatial microsimulation: Assumptions

2. Target variable is dependent on the constraint variables in a way that is relatively constant over space and time <br>

3. Input microdataset and constraints are sufficiently rich and detailed to reproduce the full diversity of individuals and areas in the study region
]]

???

Want to establish some link between the phenomena that you’re studying and the demographics that you’re using as constraints..

Points 2 and 3.

For example , you might be studying cycling propensity. There’s an association between cycling and social-demographics.

But is the relationship between socio-demographics and propensity to cycle is the same in London as it is in Newcastle – if we’re simulating from survey data over a country-level.

Final point: you want census constraints that discriminate people well.

---

## Simulating behaviour practical

.pull-left[
 <br>
<img src = "img/sim_behaviour_prac.png", width = 100%, style = "position:relative; top: 40%;"></img>
]

.pull-right[
 <br>
 .small-font[
`individuals.csv : 15,189 records`
  <br>
  `-------` <br>
  spatial microsimulation <br>
  `-------` <br>
  `simulated_oac_age_sex.csv : 320,596 records`
]]
---

## Simulating behaviour practical: survey dataset

<br>

`individuals.csv`
  `15,189 records`

<table>
   <thead>
    <tr>
     <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> var_name </th>
     <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> var_values </th>
     <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> var_type </th>
    </tr>
   </thead>
  <tbody>
    <tr>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> age_band </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> a24under, ... </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> demographic </td>
    </tr>
    <tr>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> income_band </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> 11-15k, ... </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> demographic </td>
    </tr>
    <tr>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> oac_grp </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> 1,2,3,... </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> geodemographic </td>
    </tr>
    <tr>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> uk_airport </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> MAN, DSA, ... </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> preference </td>
    </tr>
    <tr>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> overseas_airport </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> TFS, EFL, ... </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> preference </td>
    </tr>
    <tr>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> satisfaction_overall </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> 1_poor, ... </td>
     <td style="text-align:left;background-color: #ffffff !important;font-size: 16px;"> preference/attitude </td>
    </tr>
  </tbody>
  </table>

???

People holidaying to destinations…

---

## Simulating behaviour practical : use case

???

There are different types of holiday destination.

---

## Simulating behaviour practical : use case

???

Different categories of people to different destinations

---
## Simulating behaviour practical : use case

.small-font[
`individuals.csv`
  `15,189 records`
  <br>
  `--------`
]
<img src = "img/geogs.png", width = 90%, style = "position:relative;"></img>

???

This is just a survey…
Imagine we’re a Leeds-based company…
Want to know how  whole of Leeds is likely to behave
Use microsimulation to get to this level
---

## Module Schedule

<table>
 <thead>
  <tr>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> session </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> wc </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> lecture </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> lab </th>
   <th style="text-align:left;background-color: #ffffff !important;font-size: 18px;"> recap </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 1 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> 28 Feb </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> wed 1100-1800 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;"> fri 1700-1800 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 2 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 7 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> wed 1100-1800 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> fri 1700-1800 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 3 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 14 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> wed 1100-1800 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> fri 1700-1800 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 4 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 21 Mar </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> wed 1100-1800 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> fri 1700-1800 </td>
  </tr>
  <tr>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 5 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> 25 Apr </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> mon 1500-1600 </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> NA - pres </td>
   <td style="text-align:left;background-color: #ffffff !important;font-size: 20px;color: #616161 !important;"> NA </td>
  </tr>
</tbody>
</table>

.small-font[
Additional drop-in sessions via Teams: <br>
-- Tuesdays 1:00pm–2:00pm and led by [Will James](https://environment.leeds.ac.uk/geography/staff/1054/dr-will-james).
]