Lab 3

Revisiting Hertzsprung–Russell diagrams

We will once again use the data from Lab 2 - use your prior model and H-R diagram from last week. As a reminder, Hertzsprung–Russell diagrams are visualizations that show the relationship between the brightness of stars and their temperatures. Review H-R diagrams here.

The dataset for this assignment can be found as a csv file in the data folder of your repository. This dataset represents data from over six thousand stars as taken from the General Catalogue of Trigonometric Stellar Parallaxes. There are only four variables in the provided dataset:

• Vmag: Apparent visual band magnitude, a measure of brightness. Don’t worry about the units for the purposes of this lab.
• Bvcol: The color of the star, which usually corresponds to the temperature of the star (generally, negative values correspond to hot blueish stars; values around 0.5 are white, and values above 1 or so are cooler orange to red stars). Don’t worry about the units for the purposes of this lab.
• parallax: parallax in arcseconds, a measure of the distance of the star from Earth
• class: the color index of the star, another measure of the temperature of a star. The stars contained in today's dataset are of the following classes, arranged in order of decreasing temperature (i.e., B class are the hottest in the dataset): B, A, F, G, K, M. (even hotter O class stars exist, but we don't have any today!)

Exercises

1. In last week's lab, you created a linear model that predicts visual band magnitude with B-V color, parallax, and color index, treating stellar class M as the baseline/referent category. Comprehensively evaluate the linear model assumptions for this model, supported by any plots necessary.
2. Color points in your residual plot based on the color index of the relevant star. From visual inspection of your residual plot, do you notice any patterns? Describe what you notice.
3. Do you have any concerns regarding any potentially influential points? Explain, including any supporting evidence as needed.
4. Do you think that colinearity might be a concern between B-V color and parallax? Explain, including any supporting evidence as needed.

There should only be one submission per team on Gradescope. All team members must make at least one meaningful commit to the repository for this week's lab.