HimanshuMahale

Final-year undergraduate in Economics at the Indian Institute of Technology, Kanpur.

Applying to graduate programmes in economics for 2027

About

Hi, I’m Himanshu! Welcome to my website. I am currently pursuing a Bachelor of Science in Economics at the Indian Institute of Technology, Kanpur.

I am interested in understanding the economic forces that shape inequality, development, and public policy. My academic interests lie primarily in macroeconomics and development economics, particularly in questions surrounding poverty, food security, and the distributional consequences of technological change. I am especially curious about how the ongoing AI revolution may reshape labour markets, economic opportunities, and inequality.

At IIT Kanpur, I have worked on research examining the effects of Trade Policy Uncertainty and Geopolitical Risk on financial markets, using econometric methods and machine learning. I have also explored the economics of last-mile public-service delivery, studying how vehicle capacity and working-time constraints affect the cost of reaching underserved communities.

Beyond academics, I have explored the applications of quantitative methods through internships in quantitative research at Quadeye Securities and data science at American Express. These experiences have allowed me to work with optimization, financial data, and generative AI while developing a deeper interest in the intersection of economics, technology, and decision-making.

Outside academics, I enjoy being involved in campus activities and organizing book club events. I am always interested in conversations about economics, technology, public policy, and the ways in which research can help us better understand the world around us.

Research interests

Development economics
Poverty, food security, and how public services reach underserved communities.
Macroeconomics
Monetary policy, uncertainty, and how policy shocks move through the economy and financial markets.
Inequality
The forces that shape the distribution of income and opportunity.
Technology and labour
How AI may reshape labour markets, economic opportunity, and who gains from growth.

Research

Whose words move rates? RBI MPC minutes and Indian yields

Ongoing, since August 2026. With Prof. Sanjiv Kumar, IIT Kanpur

Do the individual views recorded in the RBI Monetary Policy Committee minutes move Indian interest rates? I parse every MPC statement since 2016, score each member’s hawkishness with a language model, and separate policy shocks from RBI information effects.

Details and findings
  • Scraped all 62 RBI MPC minutes from 2016 to 2026 and parsed them into 372 member statements with votes, in a tested Python pipeline that rebuilds every result from raw RBI, FBIL and FRED data.
  • Scored each statement’s hawkishness with a name-masked LLM that matches the member’s actual vote in all 372 cases and flags the 40 formal dissenters with an AUC of 0.98, benchmarked against dictionary and VADER scores.
  • Separated policy shocks from information effects using Gürkaynak–Sack–Swanson factors and Jarociński–Karadi sign restrictions on OIS and T-bill yields; 42% of decisions carry an information effect.
  • A 1-point hawkish gap between members and the Governor lifts the 1-year OIS rate by 2.3 bp on minutes day (p = 0.015), driven by internal RBI members, with null placebos but a weaker effect outside 2020–21.

Effect of trade policy and geopolitical risk on Indian markets

January to April 2026. With Prof. Sanjiv Kumar, IIT Kanpur

Does trade policy uncertainty or geopolitical risk move volatility in Indian equity markets? Using 321 months of Nifty 50 returns, I compare GARCH-X, ARDL, SVAR and time-varying models with machine-learning forecasts, and find that neither shock adds much once volatility persistence is accounted for.

Details and findings
  • Assembled a 321-month panel of Nifty 50 returns, news-based uncertainty indices and CEIC macro series, splicing three IIP base years into an HP-filtered output gap and validating outliers against RBI data.
  • Derived and coded a GARCH-X(1,1) likelihood in Python with Hessian-based standard errors and multi-start checks: persistence of 0.94 and no joint gain from TPU and GPR in a likelihood-ratio test (p = 0.48).
  • ARDL bounds tests, a Cholesky SVAR and a Kalman-filter TVP model find a long-run link via US monetary uncertainty (F = 6.75) but no significant TPU effect in any month.
  • LASSO, Ridge, Random Forest and XGBoost lose out of sample to a simple AR(1) (R² 0.79 vs 0.77 and 0.38); Diebold–Mariano tests and SHAP show ML adds no signal beyond persistence.

The cost of rural public-service delivery

September to December 2025. With Prof. Sanjay Choudhari, IIM Indore

How do vehicle capacity and working-time limits shape the cost of reaching remote villages? I replicate a vehicle routing model for 177 delivery locations, trace cost-capacity curves under different time limits, and use Shapley values to show who bears the cost of the network.

Details and findings
  • Replicated a published vehicle routing model in Python for 177 delivery locations; algorithmic optimisation beats judgment-based routing by 15.2%, saving about Rs 1 lakh a year in delivery costs.
  • Cost-capacity curves under 8, 10 and 12-hour limits show that a Rs 20,000 truck retrofit pays back in 6.7 months with no time limit but saves nothing under a strict 10-hour rule.
  • Exact Shapley cost shares for all 177 villages show that the most remote 20% bear 33% of total delivery cost, with a 0.72 rank correlation between each village’s cost and its distance.
  • Results hold under rate, fuel, seasonal and demand shocks and a noisy distance matrix; retrofit payback stays between 4.8 and 9.1 months.

Experience

Data Analyst Intern, American Express

May to July 2026

Global Risk and Compliance, Gurugram

  • Designed a quasi-experimental test-versus-control framework to measure how third-party fraud changes the spending of 100K+ customers before, during and after a fraud event.
  • Built 12 metrics and 3 new indices in SQL, including a disengagement score that compares each customer’s post-fraud spending with their own pre-fraud baseline.
  • Modelled recovery timelines with a normalised slope-based method, measuring how quickly spending returns to its pre-fraud level.
  • Estimated a ~$37M opportunity cost from fraud-driven disengagement over six months and identified a portfolio segment with a recurring loss of ~$7.5K a month.

Quantitative Research Intern, Quadeye Securities

Worked with optimization methods and financial data.

Projects

Ujjwala: access versus sustained LPG use in India

October to November 2026

Capstone project with Prof. Vimal Kumar, IIT Kanpur

A policy evaluation of India’s Ujjwala LPG scheme from 2016 to 2026, drawing on 70+ audit, parliamentary and survey sources. 10.58 crore new connections led to just 4.71 refills per household a year; low use tracks free fuelwood, cash-poor households and a connections-only target. We rank five subsidy reforms, cost four scenarios, and design a phased district rollout that doubles as an evaluation.

Monetary and fiscal policy mix in India

September to October 2026

Course project with Prof. Wasim Ahmed, IIT Kanpur

A reproducible pipeline on RBI, MoSPI and Union Budget data from 2004 to 2026. Benchmarks the repo rate against three Taylor rules, finding heavy smoothing and a 2022 tightening that lagged the rule by 3 percentage points; measures pass-through to lending rates (82% to new loans, 56% to existing ones); and shows the link between public capital spending and growth fades within two years.

Applied microeconometric models

August to November 2025

Course project with Prof. Arshad Rahman, IIT Kanpur

Hand-coded Newton–Raphson maximum likelihood in MATLAB for probit, logit, Poisson, Tobit and Heckman selection models, finding significant selection into work (ρ = −0.51). In panel models of US state murder rates, the deterrent effect of executions loses significance once Texas is dropped.

Education

Bachelor of Science, Economic Sciences

2023 to 2027

Indian Institute of Technology, Kanpur. GPA 8.2 / 10. Minors in Computer Systems and English Literature

  • 99.7th percentile in JEE Main 2023, among 1.2 million+ candidates.
  • All India Rank 6763 in JEE Advanced 2023, among 180,000+ candidates.
  • Second place among 2,500+ participants in a national technology competition.

Leadership

Editor, Vox Populi

May 2025 to April 2026

Writing and investigative journalism society, IIT Kanpur

Led a 60+ member student journalism team with 12+ editors and directed an investigation into academic inbreeding, interviewing 15+ professors. Our coverage of the 2024 student protests helped secure a monitoring committee, and readership grew to 30,000+ in a single quarter.

Book club, IIT Kanpur

I organize book club events on campus.

Contact

Email is the best way to reach me. I am happy to talk about research, graduate study, or a good book.