house_market

House prices data analysis and visualisations

House market data source: ONS. UK House Price Index.

Latest data release: UK House Price Index: September 2023 Monthly house price inflation in the UK, calculated using data from HM Land Registry, Registers of Scotland, and Land and Property Services Northern Ireland. https://www.ons.gov.uk/economy/inflationandpriceindices/bulletins/housepriceindex/august2023

Aim

Using Pablo-tester account build some charts and analysis of the Housing market using R Git and GitHub

Analysis plan

Load in data into R from xls files downloaded from ONS website
Conduct some exploratory analysis using line charts and scatter plots with simple models to assess relationships between variables
Test several univariate time series model
Automate the analysis using Targets to build a pipeline to produce a final report
Output a Quarto report highlighting the main findings in the house market data downloaded from ONS website.

All this analysis will be done in a dedicated R project using Git and GitHub for version control, adding new analytical features as new branches to main using my local repo as Pablo-tester.

Using trunk development to add new features

In this project I use trunk-based development to add features to the main branch. Pablo-tester GitHub account creates the features and Pablo-source GitHub accounts reviews and approves the changes applied. Git_commit_branch_ahead

Adding chages from Git to GitHub

Close_issue_from_terminal

Closing issues as I push the fix, referencing issue number in commit message

1. Exploratory data analysis

1.1 Average house price change (%)

UK Average House price 2005-2013 01_Average_UK_house_price_2005_2013

UK Average House price Time Series decomposition (seasonality, trend and random):

05_Average_UK_house_price_TS_decomposition_ADDITIVE

Year on year and month on month UK Average house price change show negative rates for first time in September 2023 since 2011. Included reference point (latest value) for each series (MoM and YoY percent change) by adding these to lines to 04_House_price_change_plots.R script:

'endv <- group_by(Price_change_labels, metric) %>% filter(datef == max(datef)) geom_point(data = endv, col = 'darkgray') + geom_text(data = endv, aes(label = percent), hjust = -0.4, nudge_x = 2)'

03_UK_MoM_and_YoY_price_change_DEC2023_env_MARK

2. Univariate TS modelling using TSstudio

We will explore how to create, train and compare several TS univariate models using TSstudio, using UK Average House price data.

The TSstudio package provides a set of tools descriptive and predictive analysis of time series data. That includes utility functions for preprocessing time series data, interactive visualization functions based on the plotly package engine, and set of tools for training and evaluating time series forecasting models from the forecast, forecastHybrid, and bsts packages.