1

I hypothesise that the higher the outdoor temperature, the higher the surface temperature of the building envelope, which should also translate into the detection of higher surface temperature on the other side of the surface (eg: temperature measured on the outer vs inner side of a wall). I have collected temperature data over a time period with approximately 300 readings. However, I am a little lost over which model is appropriate to study such a relationship. Would mediation analysis be appropriate?

Any feedback would be most appreciated as I am not very well-versed in statistics.

user401174
  • 11

1 Answers1

1

I don't see any mediation happening here. Mediation happens when one variable "get in the way" of another relationship. A classic example is that the more firefighters go to a fire, the more damage is done.

Does this mean we should send fewer people to fight fires?

No, the relationship between firefighters and damage is mediated by the size of the fire. Bigger fires get more firefighters and do more damage.

You seem to have several linear regression: Higher outdoor temperature makes everything hotter.

As to type of analysis, I'm guessing some kind of time series, but that is not my area of expertise.

Peter Flom
  • 119,535
  • 36
  • 175
  • 383