Condition-based maintenance of machinery is being much talked about in the engineering sector of defense and commercial industry. A lot of expenditure is generally incurred on condition monitoring of machinery to avoid unexpected downtimes and failures vis-à-vis optimizing machinery operation. The concept is ever-evolving due to technological advancements as well as with the emergence of the unique nature of defects. Vibration Analysis is a potent tool of condition monitoring for prediction and diagnostics of machinery failures. Presently, time and frequency spectra are being widely used for defect diagnostics of machinery. However, they require signal conditioning to eliminate noise and to enhance the resolution of the spectrum. Extensive research in the area of signal processing has been undertaken to refine time and frequency spectra. Notwithstanding the application of statistical tools for analysis of various defects in machinery using condition monitoring, data can be a viable option. Research in this area, where statistical models have been applied, revealed encouraging results. In this paper, we have modeled bearing vibration data by applying the time-varying Markov Switching Auto-Regressive method which was found very helpful in estimating RUL of machinery.