Scribd
Upload
61 views7 pages

Super Forecasters: Project Proposal Pitch

This document provides a proposal for a project to predict crime rates. It will collect data on crime rates, demographics, economic activity, employment rates, geographic attributes, and time trends from various government sources. It will analyze the data using LASSO to determine the key variables correlated with crime. It will then use these variables to predict where and when future crimes may occur. Visualizations and a dashboard with heat maps will also be developed. The timeline outlines data collection over 4-6 weeks, analysis from weeks 6-8, and dashboard development from weeks 8-10.

Uploaded by

daselknam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
61 views7 pages

Super Forecasters: Project Proposal Pitch

This document provides a proposal for a project to predict crime rates. It will collect data on crime rates, demographics, economic activity, employment rates, geographic attributes, and time trends from various government sources. It will analyze the data using LASSO to determine the key variables correlated with crime. It will then use these variables to predict where and when future crimes may occur. Visualizations and a dashboard with heat maps will also be developed. The timeline outlines data collection over 4-6 weeks, analysis from weeks 6-8, and dashboard development from weeks 8-10.

Uploaded by

daselknam
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 7

Super Forecasters

Project Proposal Pitch


Andrew Mao, Faraz Ahmed, Pedro Armengol
Agenda

Problem/Goal

Data Collection/Manipulation

Analysis/Forecasting/Visualization

Timeline
The Problem/The Goal

Crime rates:
heterogeneous

Which are the


factors that are
correlated to
crime?
With those
factors, can
we predict
crime?

Goal: Find out the set of variables most closely correlated with crime,
and predict with a certain level of accuracy where and when
next crime will take place.
Data Collection
Sample data collection sources:
Predicted Predicting 1)Crime data: monthly/municipality level:
variables variables National System of Crime Data
Sample
Independent 2)Demographic information:
-Economic
Variables yearly/municipality level:
activity National Statistics Bureau (NSB)
-Cost of
living 3) Employment rates: yearly/municipality
-Population level: National Insurance Institute
-Education 4) Geo. Attributes: municipality: NSB
Crime rates rates 5) Economic activity/Cost of living:
-Employment
rate quarterly municipality:
-Geo. Economic Ministry
Attributes Data cleaning Procedure:
-Time trends 1) Scrapping data
2) Standardize data to same units
3) Design the database/tables (SQL)
4) Sort/merge/link (relational database)
Analysis

Covariate Analysis

LASSO (Least absolute shrinkage and


selection operator)

Heat Map
Visualization/Dashboard

Prediction & Forecasting


Timeline

Gather data
4th – 6th Week Clean collected data
Standardize and merge data

Examine patterns, perform


6th – 8th Week analysis and apply forecasting
techniques

8th – 10th Week Develop the dashboard (with


heat maps)
Thank you for listening!

Questions are welcome

You might also like