Stellar demographics, the occurrence rate of stars, is studied to confirm theories of star formation and stellar evolution. The number of stars existing right now can be looked upon as the consequence of the rate at which they are being created at each possible type (basically, by mass, modeled by initial mass functions, IMFs) and expected lifetime of each type. Consistency between these three can be checked using occurrence rates from surveys, taking into account survey biases, such as the Malmquist bias and Eddington bias. In citing (or reading) percentages/fractions of occurrence, it is necessary to be clear about the term "star", which in various contexts, may or may not be intended to include brown dwarfs and/or each type of stellar remnant.
See "spectral class" for abundances by class, which make it clear that of main sequence stars, 99% fall in the F-type to M-type range, with O-type through A-type accounting for less than 1% (though they are so bright that they often can be seen from distances at which cooler stars cannot). Some post-main-sequence phases are also extremely bright and also comparatively short-lived, accounting for on the order of a percent or less of current stars.
Among stellar remnants, white dwarves do shine but are so small that only nearby ones can be seen. Over the long term in the future, the number (and population fraction) of stellar remnants will presumably grow, as will the number of M dwarves, which have lifetimes hundreds to thousands of times longer than stars like the Sun, assuming that future generations of stars continue to be created according to a similar IMF.