# Regression response and explanatory variables

I am a newbie to regression, and I was trying to answer the following question using this data.

Is there a meaningful difference between the distribution of damage caused by hurricanes with female names and the distribution of damage caused by hurricanes with male names?

The response variable is the focus of a question in a study or experiment. An explanatory variable is one that explains changes in that variable.

By using the definition, gender affects the distribution of damage caused by hurricanes, and it's the explanatory variable. Am I right?

By considering gender as an explanatory variable, does GAM regression give a good fit?

set.seed(100)
library(mgcv)
library(rio)
data = rio::import(url)
data = data[1:92,]
log.reg = gam(NDAM ~ Gender_MF, data = data)

• You should consider using a logistic regression when the outcome (dependent, left-hand-side, etc.) variable is binary. When the regressor (right-hand-side variable, explanatory variable, etc.) is binary you can do linear regression. lin.model <- lm(NDAM ~ Gender_MF, data = data) Yours is the latter case, so logistic regression is unnecessary. Jul 18, 2019 at 3:40
• @doubletrouble I appreciate your explanation. My confusion is on the question, "Is there a meaningful difference between the distribution of damage caused by hurricanes with female names and the distribution of damage caused by hurricanes with male names?". From the question, gender is an explanatory variable. Is that right? Jul 18, 2019 at 3:46
• I've addressed your first question below, but your second question is unclear. Logistic regression would be an appropriate way to analyse the data if gender was the response variable, but what you do mean by 'give the best fit'?
– mkt
Jul 18, 2019 at 9:22
• It seems to me that there may be some confusion in terms. I do not agree that the term "explanatory variable" is defined as "one that explains changes in that variable." In regression analysis, there is no need to have an explanatory varible that explains the change in a variable. An explanatory variable in this context is simply variable (on the RHS of a regression equation) that is thought to explain some of the variability in the response variable. There needs to be no causal effect of the explanatory variable. Regression is often carried out to ... Jul 19, 2019 at 15:11
• (continued) determine if an explanatory variable does in fact have account for some of the variability seen in the response variable. Jul 19, 2019 at 15:12

You are correct: gender would be an explanatory variable (or predictor/independent variable) in this case.

However, you should be aware that the paper you are presumably basing this on has received an extraordinary amount of criticism for poor analysis. For e.g.

https://www.washingtonpost.com/news/monkey-cage/wp/2014/06/05/hurricanes-vs-himmicanes/?noredirect=on&utm_term=.81faaced2590

https://scatter.wordpress.com/2014/06/03/my-thoughts-on-that-hurricane-study/

https://scatter.wordpress.com/2014/06/10/the-hurricane-name-study-gets-worse/

https://www.sciencedirect.com/science/article/pii/S2212094715300517

http://rpubs.com/oharar/19171

Not to mention 4 published responses to the paper itself: https://www.pnas.org/content/early/2014/05/29/1402786111