Happy to announce that I would join the National University of Singapore (NUS) as an Assistant Professor in Fall 2021. I have a PhD and a Postdoc opportunity starting from Fall 2021. Please do reach out for further enquiries. #academia #singapore #phdchat #highereducation

Thanks for sharing our work @jfulponi

Got a chance to share my thoughts on the future of car-sharing in Singapore at CNA's Talking Point. I mainly talk about the economics of carsharing from the consumer's perspective. youtube.com/watch?v=cjkQEELn… #CNA #economics #carsharing #singapore

Please check out our new paper on a robust discrete choice model in "Statistics and Computing". Fat-tails of t-distribution make this model less sensitive to outliers: link.springer.com/article/10… #bayesian #robustness #discretechoice #statistics

The recent paper that I like a lot! A new dynamic model for crowding valuation in subway/metro. As one of the reviewers said -- a good illustration of how problem-driven statistical models can be developed. #statistics #crowding #valuation #dynamicmodels #behaviour

How can we use vehicle location data, unsupervised machine learning, and simulation techniques to detect primary and secondary disruptions in metro systems? #publictransport #metro #disruption #simulation #MachineLearning sciencedirect.com/science/ar…

Starting my "Neurophysiology and Behavioural Science Lab" in Civil Engineering at NUS. We have: eye-tracker EEG VR headset GSR ECG Is it not truly interdisciplinary? More details are awaited... #neuroscience #behaviouralinsights #neurophysiology #Psychology

Our new paper in Energy Economics highlights the sensitivity of Indian and Chinese car buyers relative to demand-side policies! Main result: Demand-side policies are ineffective in reducing fuel consumption. A supply-side policy is a way forward. #energy #energytwitter

How to model animal-vehicle collisions using large-scale data and find the effect of countermeasures on AVC reduction? How to account for unknown animal-vehicle encounters & seasonality? with @KaraKockelman & co-authors #safety #bayesian #AnimalCrossing sciencedirect.com/science/ar…

How can you mess up so much in one journey: 1. Delayed connecting flight by 24 hours 2. Delayed baggage by 48 hours 3. Serving non-vegetarian meals even after requesting vegetarian mean Least you can do is to respond my request (FB ID: 35722376) @lufthansa @lufthansaNews

Can a rebate/fee on upfront price based on fuel economy be an effective fuel conservation policy in India & China? No, we answer it by modelling vehicle choice and usage using revealed preference data: arxiv.org/abs/2201.08995 #China #India #environmental #BehaviorTwitter

Can aggregated mobility data help in predicting spread of pandemic? Please check out our new compact matrix-based model. Studied first wave of COVID-19 infections in Italy using mobility between provinces. papers.ssrn.com/sol3/papers.… (preprint) #Epidemiology #covid #bigdata

As soon as I joined a tenure track position, time seems to fly much faster during the day. #AcademicTwitter #AcademicChatter

Is 'building more roads' a solution to congestion? Check out our latest preprint where we answer this question by quantifying the nature and form of congestion in urban road networks. papers.ssrn.com/sol3/papers.… #congestion #cities #bigdata #bayesianstatistics #machinelearning

E-bikes are emerging, but little is known about how to include them in design of shared sustainable mobility infrastructure. @khashayar1988 & I discuss lessons learned from lit review on e-bike LOS. sciencedirect.com/science/ar… #emobility #ebikes #bikes #sustainabletransport

We reject two myths that i) mixed logit models with inter (across individuals) and intra (across time) heterogeneity predicts consumer preferences better and ii) its MCMC estimator is superior to likelihood-based. sciencedirect.com/science/ar… #mixedlogit #econtwitter #consumers

New preprint: General framework to forecast the adoption of novel products (e.g., autonomous vehicles). Modelled word-of-mouth dissemination through social networks & its effect on risk preference. arxiv.org/abs/2109.06169 #MarketingTwitter #EconTwitter #AutonomousVehicles

Compiled codes for econometric methods developed by me & collaborators: prateekbansal.org/software Models: discrete choice, spatial count, hidden Markov Estimators: MLE, variational Bayes, MCMC, EM Languages: Python, Matlab, R, Julia #econometrics #Statistics #AcademicTwitter

Our new paper in the Econometric journal on Quadrature methods to approximate integrals in maximum simulated likelihood estimation academic.oup.com/ectj/advanc… #Econometrics #Statistics #EconTwitter #integration #simulation

New spatial count data model to tackle temporal instability: Integrated state-space model into spatial count data model to account for time-varying parameters. Derived fully conjugate Gibbs sampler. authors.elsevier.com/a/1dP~q… #spatialstatistics #bayesianstatistics #safety

Once Consumers Adopt an Electric Vehicle, Do They Go Back? Some instant insights from our preliminary analysis of a unique dataset: kapsarc.org/research/publica… #ElectricVehicle #energy #energytwitter #consumers #behaviour #environment #Sustainable