# Techniques for dealing with unevenly spaced time series data that have missing time-stamps?

What are the techniques to infer missing time-stamps in the unevenly spaced time series data that has missing time-stamps?

The frequency of data recording is about 1-3 times a day without any fixed time of measurement. In some cases the time stamp of the measurement is unavailable.

It is possible to predict at which time $t_0$ the new sample $x_0$ was recorded knowing the history of previous measurements $(t_n,x_n)$?

$$(t_{-n},x_{-n}), ... ,(t_{-3},x_{-3}), (t_{-2},x_{-2}), (t_{-1},x_{-1}), (t_0,x_0)$$

Or predict multiple missing time-stamps e.g. $t_{-2},t_{-6},t_{-7}$ from recorded sequence $t_n,x_n$?