9 12 2016

2: Context of Project

This pitch is about a little R Shiny App which has been created as course project in the Online Course "Developing Data Products" by Johns Hopkins University at the platform Coursera. It is part of the course series "Data Science". The purpose of the app is to demonstrate the capability of using basic Shiny functionality, i.e.

  • using interactive elements
  • doing some computation on the server side
  • playing back the results interactively to the user interface.

What the app does is predicting an outcome of a linear regression model and display this prediction and a scatterplot in different units, depending on wheater the user prefer's the US system or metric units. The data set and the model is described at the following slide.

3: The Data Set

The data set 'airquality' incorporates 153 measurement points of different parameters at weather stations in New York, from May to September 1973. The variables of interest for this project are wind speed and temperature, which are used in a prediction model. The dataset comes with base R and looks like this:

head(airquality)
##   Ozone Solar.R Wind Temp Month Day
## 1    41     190  7.4   67     5   1
## 2    36     118  8.0   72     5   2
## 3    12     149 12.6   74     5   3
## 4    18     313 11.5   62     5   4
## 5    NA      NA 14.3   56     5   5
## 6    28      NA 14.9   66     5   6

The data set and its variables are documented here: https://stat.ethz.ch/R-manual/R-devel/library/datasets/html/airquality.html

4: Predicting Temperature by Wind Speed

The analytic questing is how measurements of temperature relate to those of wind speed. At the global level solar energy and its distributional pattern across the globe is the basic driver of all weather phenomen such as wind systems. But on the local level it is known that higher wind speed go hand in hand with lower temperature (at least almost everybody knows this from experience). Thus, temperature (criterion) is regressed on wind speed (as the predictor) in a simple linear regression model based on the available data points from the New York weather stations. Based on the model the user selects a certain wind speed of his or her choice and get's back the predicted temperature according to the model. In addition, the user can select wheater the computations are run based on US or on metric units.

5: Let's Start!