Creating a model is an essential part of forecasting and data analysis. I’ve put together a quick guide on my process for modelling data and checking model fit.
The source data I use in this example is Melbourne’s weather record over a 12 month period. Daily temperature is based on macroscale weather and climate systems, however many observable measurements are correlated (i.e. hot days tend to have lots of sunshine). This makes using weather data great for model building.
Before we get started, it’s useful to have some packages up and running.
#Useful packages for regression library(readr) library(readxl) library(ggplot2) library(dplyr)…
I love maps. I’m also quite a fan of R.
Over the past few months I’ve been exploring how best to create maps of Australia using R and data from the ABS. This is what I’ve learnt.
Firstly, the ABS update their geographies regularly. They can all be readily accessed here.
You’ll want to download these in ‘ESRI Shapefile format’ (e.g. Statistical Area Level 2 (SA2) ASGS Ed 2016 Digital Boundaries in ESRI Shapefile Format). SA’s or statistical areas are the ABS’ way of carving up the country into manageable regions for analysis.
Secondly, there’s a myriad of packages out…
I have always loved weather. I’m also lucky enough to live in Australia where weather data for thousands of locations across the country is recorded and uploaded by the Bureau of Meteorology every day.
When thinking about living in different towns or cities, I’ve always struggled to picture what the conditions are like on the ground — particularly at different times of the year. Sure there’s snapshots of the heatwave or the morning frost, but what about between these events?
I think I’ve figured out the best way to visualise climate variables for a year round view. …
Note on 10 April 2020: I first wrote documented my thoughts on COVID-19 on 25 February 2020. This article has been updated to reflect how the situation has changed over the past 6 weeks.
I first started seriously worrying about COVID-19 in mid February. At the time, Australia had 15 confirmed cases (all from overseas) and the infection rate was 0. That is, there were no new cases developing each day.
It has certainly been a year. Unsurprisingly, having lived through the Melbourne lockdown during 2020 (twice!), I have found more time to read than in any previous year.
I also drove from Brisbane to Melbourne round trip three times— giving ample a chance to catch up on ‘those books’ that I mention in conversation almost weekly but have actually never read (or listened to) start to finish.
Below is a list of all the books I got through in this wild year. Included are all 3 formats: paperbacks, epubs, and audiobooks.
The Man who Solved the Market — Gregory Zuckerman
With nearly one hundred bushfires still ravaging New South Wales, Victoria, South Australia and Tasmania, the Bureau of Meteorology has reported frequent record maximum temperatures from across the nation over the past 3 months.
I had been focusing on my local QLD figures, recorded at the Brisbane CBD station. My aim was to determine just how unprecedented the 2019 mean monthly maximum temperature profile had been.
I started by looking at the Brisbane story. As seen below, it’s been the hottest November on record.
However, as I am moving to Melbourne in 10 days time, I thought I would compare…
Originally published on 20 December 2019
It’s 20 days into December, and one thing has become clear. This weather is not normal. November 2019 was the hottest on record for Brisbane, and December will almost certainly achieve the same feat.
The Bureau of Meteorology provides access to daily temperature data for each weather station across the country. Downloading the Brisbane data set, it is obvious that temperatures in 2019 are at the top of the mean maximum temperature range for almost every month.
With the ten hottest days of summer still to come, it seems the 2005 record temperatures will…
2019 was the first year I worked full time as a professional Economist. The two main industries I provided advice on were science and education. I’ve taken some time to distil my experience into five themes that shaped the year, with the aim to track how my perspective changes over time.
Theme 1: You are often left to your own devices, and more often left to your own assumptions.
Assumption setting is one of the main expertise areas of an economist, however I have been surprised by how few assumptions had been developed [normally by government] when analysing a problem…
[Originally published August 14, 2018]
China’s commitment to addressing the causes and effects of climate change span well beyond their fair share of responsibility and should be used as a model for other developed economies.
China is committed to peak the nations carbon dioxide emissions by 2030 at the latest, while reducing CO2 intensity by over 60% from a 2005 baseline. While these targets put China at the front of the pack for renewables and transitions targets, it is their role in research and development into PV solar panels that create positive externalities and technical adoption opportunities for economies.
In the 2016 Australian Federal Election, over 720,915 people (5.5% of all votes cast) voted informally. Of these, over half (377,585) had ‘no clear first preference’, meaning their vote did not contribute to the campaign of any candidate. This raises 3 main questions:
The answer to each of these three questions yielded highly unexpected results.
What does it mean to lodge an ‘informal’ vote?
Short answer: A vote is deemed ‘informal’ if it is not filled out in accordance…