Make sure the time the device records the data is tracked along with the time the data is received. You can monitor delivery times to diagnose issues and keep a close eye on information lag times. For example, if you notice the delivery time steadily increases just before you get a data loss, it is probably something in the networking and not the device that is causing the issue.

Since MQTT enables subscribers and publishers to be offline if necessary, the time the message is received could be very different from when it was originated. This can affect the predictive modeling accuracy if the time received becomes significantly distant from when the event being analyzed actually occurred. If connections are persistent, this should be a non issue.