The time-averaged shear wave velocity of top 30 m (V_S30) is the most commonly used parameter to classify a site and evaluate its amplification characteristics for the seismic design. The in-situ seismic tests must be performed up to a depth of 30 m for obtaining the shear wave velocity (V_S) profiles to estimate V_S30. It is intimated that, in most of the cases, the measured V_S profile does not extend up to 30 m due to numerous reasons including limitation of testing techniques and unfavorable field conditions. Since, the measurements of V_S30 are unavailable for the majority of Pakistan and the world, the local geology and topographic slope or its combination are used to estimate V_S30. However, there is no field-based validation of the estimated V_S30 is performed in Islamabad–Rawalpindi region, the proxy-based estimation may lead to unrealistic results. To accommodate this, region specific extrapolation methods are developed. This study develops an empirical data-driven function of V_S30 from shallow V_S profiles by correlating V_S30 with the time-averaged V_S to depths less than 30 m. In this regard, 85 V_S profiles are used from Rawalpindi-Islamabad region. A comparative analysis of the proposed procedure is carried out with the published methods. It is revealed that V_S30 predicted by the proposed function results in close matches with the data measured in the western United States. In addition, the results indicate that the local geology and topographic slope proxies may not be acceptable for usage in the region due to their greater uncertainty. Finally, a procedure for extrapolating the V_S profile from available shallow depth measurements up to 30 m is proposed.