• Users will be able to input predictor values of their choosing
• Once input values are set, users click a submit button
• Once values are submitted, the user receives predicted Ozone level as output

The following script displays the linear model used to predict Ozone.

data(airquality)
    airquality <- subset(airquality, select = -(Day))
    fit <- lm(Ozone ~ ., data = airquality)
    summary(fit)$coefficients

##                 Estimate  Std. Error   t value     Pr(>|t|)
## (Intercept) -58.05383883 22.97114118 -2.527251 1.297198e-02
## Solar.R       0.04959683  0.02345827  2.114258 3.683836e-02
## Wind         -3.31650940  0.64578918 -5.135591 1.286145e-06
## Temp          1.87087379  0.27363201  6.837189 5.337230e-10
## Month        -2.99162786  1.51592267 -1.973470 5.104492e-02

• Increased understanding of how certain variables may affect Ozone
• Interactivity with multiple linear regression
• Exposure to the layout of a shiny application

• Delayed reactivity by using submit button
• Multiple linear regression model creation
• Use of sliders to make it easier for the user to input values
• Multiple tabs on website to differentiate app from documentation