Hsma 8b Forecasting

Due to some environment issues that have not been fully resolved, you may run into issues on some platforms when running the code. This is due to some changes in newer versions of packages that were necessary to include for compatability reasons. For now, I would recommend just watching the lecture videos. However, if you are keen to still try the exercises/code-alongs but run into issues, I would recommend skipping to the next part of the exercise/code-along notebook or the next exercise/code-along notebook. For all code-along notebooks and exercises, you have several options. Run it locally on your machine using the hsma_forecast environment

Exercises created by Tom Monks, with some modifications by Elliot Coyne. Monks, T. (2023). forecasting health service demand in python. Zenodo. https://doi.org/10.5281/zenodo.4332600

There was an error while loading. Please reload this page. In this session, we spend a bit of time learning about Quarto, allowing you to create good-looking reports that weave together code, text, images and more in a neat, easily-distributable format. We also have a look at using a Python script to interact with the command line and produce multiple outputs from a single Quarto file using different parameters. We then briefly dive into the world of xlsxwriter - a handy tool for automatically producing Excel spreadsheets from your Python code. This session makes use of the hsma_reproducible_reporting environment

Time-series forecasting methods are a useful thing to have in your analyst toolbox. In this session, we cover how to prepare data for forecasting, how to train simple models, how to assess your forecasts, and how to use the modern forecasting library Prophet. We also discuss when time series forecasting methods are a good fit for your problems, and when you should consider turning to other approaches. This session has not yet been fully rewritten for HSMA 6. However, you can watch the sessions from HSMA 5, which cover the same material. Some bonus videos from previous rounds of HSMA are also linked below.

Due to some troublesome environment issues, the exercises are not currently completely functional. We would instead recommend Tom Monk's excellent exercises, on which the HMSA 5 exercises were originally based. Tom's exercises have been updated to work in 2024 and are used on the Exeter Health Data Science MSc. In this slide-only session, we go over some useful things you may want to consider in your own coding projects, as well as briefly touching on a range of topics that we couldn't cover... It's useful to know of these approaches so you are aware of the possibilities and can spot situations in which these may be a more appropriate approach than the ones we've had time to... Using your Python knowledge and general data science and operational research concepts, you are now well-placed to dig further into any topics in this session that catch your eye.

Within the organisation (NHS England) forecasting for clock starts is currently done using scenario-based modelling in Excel. This limits how much data can be used to make forecasts, and the techniques that can be deployed. There was a desire to explore other forecasting methods to see if they could obtain more accurate forecasts, and also to valid the forecasts that are being made using the existing model. Understanding demand is important to manage wait lists, which is a high priority for the NHS, hence the interest in modelling clock starts. The aim of the project was to explore different ways of forecasting demand, evaluate their performance and offer suggestions as to how forecasting could be done in the future. The team carried out exploratory data analysis to understand differences between different data types and the differences in the patterns of demand for different subgroupings.

They built a codebase that loaded the required data and ran model functions for Naïve, ARIMA, Prophet and a combination of Linear Regression and Random Forest. They evaluated the performance of these models to each other, and the modelling technique currently used. They also made some progress on rebuilding the current excel model in Python. They found that while the more performant models that we used performed well (produced low error metrics), they generally didn’t perform as well as the model currently used within the organisation. This suggested that a helpful next step of the project might be to complete the work of translating the excel model into python and doing further work to understand if they can get more... There was an error while loading.

Please reload this page.

People Also Search

• HSMA 8B - Forecasting
• GitHub - hsma-programme/8b_forecasting
• 2_naive_models_SOLUTIONS.ipynb - Colab
• 8B: Time-Series Forecasting - HSMA
• PDF 8b_forecasting/HSMA 5 Slides - Part 2 - Forecasting with ... - GitHub
• 03_benchmark_forecasts.ipynb - Colab
• Module 8: Modern Analytics - HSMA
• HSMA Project Handbook - Forecasting Demand - Investigating approaches ...
• 3_forecasts_with_prophet_SOLUTIONS.ipynb - Colab
• 8b_forecasting/index.qmd at main · hsma-programme/8b_forecasting