Abstract
Bank shares are the most actively traded equities in Uzbekistan and the main channel through which the state is expected to sell its stakes in banks. This paper tests the weak form of the efficient market hypothesis for the ordinary shares of four listed banks – Uzpromstroybank (SQB), Hamkorbank, Ipak Yuli Bank and Aloqabank – and for the UCI index, using 950 daily closing prices from January 2023 to September 2026. Autocorrelation tests, Lo–MacKinlay variance ratios with robust statistics and the Chow–Denning multiple test are applied separately to 2023–2025 and to January–September 2026. In 2023–2025 three of the four bank stocks show strong next-day reversals (first-order autocorrelation from −0.29 to −0.35; ten-day variance ratios of 0.26–0.40), and the random walk is rejected at the 1% level. In 2026, when the monthly trading value of the two most traded bank stocks rose more than tenfold, no series rejects the random walk: Hamkorbank’s returns become close to white noise, while SQB switches from reversal to short-run momentum. Thin trading appears to be the main source of inefficiency, and single-day quotes are a poor benchmark for pricing sales of state-owned bank shares.
References
This work is licensed under a Creative Commons Attribution 4.0 International License.