http://dspace.iimk.ac.in:80/xmlui/handle/2259/929 2026-05-12T13:16:12Z 2026-05-12T13:16:12Z Mukherjee, Soumyatanu http://dspace.iimk.ac.in:80/xmlui/handle/2259/941 2020-12-08T17:23:17Z 2017-03-01T00:00:00Z

Exchange Rate Volatility and Exports: Estimation of Firms Risk Preferences Mukherjee, Soumyatanu In this companion paper to Broll and Mukherjee (2017), we empirically analyse how exchange rate volatilities a ect rms optimal production and exporting decisions. The rms elasticity of risk aversion determines the direction of the impact of exchange rate risk on exports. Based on a exible utility function that incorporates all possible risk preferences, a unique structurally estimable equation is used to estimate the risk aversion elasticities for a panel of Indian service sector (non- nancial) rms over 2004-2015, using the quantile regression method.

2017-03-01T00:00:00Z Villari, Boby Chaitanya Abdulla, Mohammed Shahid http://dspace.iimk.ac.in:80/xmlui/handle/2259/935 2020-12-08T17:23:11Z 2017-01-01T00:00:00Z

effSAMWMIX: An efficient Stochastic Multi-Armed Bandit Algorithm based on a Simulated Annealing with Multiplicative Weights Villari, Boby Chaitanya; Abdulla, Mohammed Shahid —SAMWMIX, a Stochastic Multi-Armed Bandit(SMAB) which obtains a 𝑶𝑶(𝒍𝒍𝒍𝒍𝒍𝒍 T) where T being the number of steps in the time horizon, is proposed in the literature . A blind-SAMWMIX which incorporates an input parameter ,which has better empirical performance but obtains a regret of the order 𝑶𝑶(𝒍𝒍𝒍𝒍𝒈𝒈𝟏𝟏+𝟐𝟐𝜶𝜶 𝑻𝑻).Current work proposes an efficient version of SAMWMIX which not only obtains a regret of 𝑶𝑶(𝒍𝒍𝒍𝒍𝒍𝒍 K) but also exults a better performance. A proof for the same is given in this work. The proposed effSAMWMIX algorithm is compared with KL-UCB and Thompson Sampling(TS) algorithms over rewards which follow distributions like Exponential, Poisson, Bernoulli, Triangular, Truncated Normal distribution and a synthetic distribution designed to stress test SMAB algorithms with closely spaced reward means. It is shown that effSAMWMIX performs better than both KL-UCB & TS in both regret performance and execution time

2017-01-01T00:00:00Z Villari, Boby Chaitanya Abdulla, Mohammed Shahid http://dspace.iimk.ac.in:80/xmlui/handle/2259/934 2020-12-08T17:22:29Z 2017-01-01T00:00:00Z

Portfolio choice decision making with NBP-effSAMWMIX: A Stochastic Multi-Armed Bandit Algorithm using Naïve Bandit Portfolio Approach Villari, Boby Chaitanya; Abdulla, Mohammed Shahid Portfolio Selection Problem (PSP) is actively discussed in financial research. The choice of available assets poses the need for exploration and the objective to maximize the portfolio payoffs makes the PCP an explore-exploit decision-making problem. Multi-armed bandit algorithms (MAB) suit well for such problems when applied as the decision engines in Naïve Bandit Portfolio algorithms (NBP). An NBP’s performance varies by varying the MAB inside the algorithm. In this work we test a Stochastic Multi-Armed Bandit (SMAB) named effSAMWMIX, which we proposed in a previous work of ours, to solve the PSP. We compare the performance of effSAMWMIX vis-à-vis KL-UCB, Thompson Sampling algorithm and the benchmark Market Buy & Hold strategy. We tested the algorithms on simulated and real-world market datasets. We report our results where effSAMWMIX, applied as the decision-making engine of NBP, has achieved better cumulative wealth for all portfolios when compared to the competing SMAB algorithms.

2017-01-01T00:00:00Z Gangopadhyay, Kausik http://dspace.iimk.ac.in:80/xmlui/handle/2259/933 2020-12-08T17:22:26Z 2017-01-01T00:00:00Z

A Survey into Evidence of Zipf’s Law among Indian Socio-Economic Variables Gangopadhyay, Kausik Zipf’s Law is an empirical phenomenon observed in many natural systems. The distribution of a physical variable demonstrates sharp rise at the right tail under this law. The occurrence of this law is pervasive among the physical variables. Econophysics, a discipline named so by Eugene H. Stanley, studies the application of physical principles among variables related to human action mostly related to socio-economic variables. This paper surveys the studies on the existence of Zipf’s law among Indian socio-economic variables. We present the evidence on economic variables, in particular income, wealth and consumption distribution. The other socio-economic variable of our choice is city size distribution. In all cases, the Zipf’s law is established with different values for the Pareto exponent.

2017-01-01T00:00:00Z