Hi, my name is

Amin.

I am a Data Scientist and ML Engineer.

My focus is on theories and applications of modern statistics across domains including Finance, Business, Health and Traffic.

Hugo Profile profile image

About Me

Currently I am a Ph.D. student in Business Analytics at the University of Cincinnati, Lindner College of Business. For my undergraduate degree, I pursued Financial Management at the University of Tehran, Faculty of Management. My research focus is on empirical asset pricing using machine learning. My research also involves theoretical and empirical properties of machine learning and deep learning methods, as well as the advantages and disadvantages of ML compared to conventional statistical methods.

In addition to my research, I have been also involved in teaching and mentoring students since 2021 with more than 2K in-person and online students in Iran and United States in total. I have more than 100 hours of recorded class videos in this area (See teaching section). I enjoy explaining quantitative methods in economics, econometrics, and finance, with an emphasis on building intuition behind formal models and bridging theory with practical applications using Python programming language.

Also, I have a long-standing interest in chess, which has shaped my analytical thinking, and I value discipline and consistency through regular physical training.

Here are a few keywords of my research I work on:
  • Empirical Asset Pricing
  • Market Efficiency
  • Explainable AI
  • Model Interpretation
  • Timeseries Analysis

Publications

A comprehensive study of market prediction from efficient market hypothesis up to late intelligent market prediction approaches

Authors: A Aminimehr, A Raoofi, A Aminimehr, A Aminimehr

Journal: Computational Economics
Volume: 60 (2)
Pages: 781-815 Year: 2022 Citations: 23

ParsBERT post-training for sentiment analysis of tweets concerning stock market

Authors: M Pouromid, A Yekkehkhani, MA Oskoei, A Aminimehr

Journal: 2021 26th International Computer Conference, Computer Society of Iran (CSICC) Year: 2021 Citations: 21

The role of feature engineering in prediction of tehran stock exchange index based on LSTM

Authors: A Aminimehr, A Raoofi, A Aminimehr, A Aminimehr

Journal: Iranian Journal of Economic Studies
Volume: 9 (2)
Pages: 527-548 Year: 2020 Citations: 8

A study on the characteristics of TSE index return data and introducing a regime switching prediction method based on neural networks

Authors: A Aminimehr, S Bajalan, H Hekmat

Journal: Journal of Financial Management Perspective
Volume: 11 (34)
Pages: 145-171 Year: 2021 Citations: 6

The strength of convolutional neural network in financial distress prediction

Authors: A Aminimehr, H Hekmat

Journal: Financial Management Strategy
Volume: 11 (2)
Pages: 77-96 Year: 2023 Citations: 2

Stock market dynamics through deep learning context

Authors: A Aminimehr, A Aminimehr, HM Kamali, S Eetemadi, S Hoseinzade

Journal: arXiv preprint arXiv:2405.09932 Year: 2024 Citations: 5

Teaching Experience

Gradute Teaching Assistant - UC
Aug 2025 - Present
I teach core business analyitcs courses to undergraduate degree students at Lindner College of Business, University of Cincinnati.
Gradute Teaching Assistant - UWM
Sep 2024 - May 2025
At University of Wisconsin Miwaukee department of Economics, I was in charge of conducting weekly review sessions to summarize class materials and solve problems for undergraduate students in Macro-economics classes. Additionally, I held regular office hours and contributed by proposing exam questions.
Course designer and instructor - Alzahra University
2022 and 2023
I designed and taught an introduction to Python programming language and financial time series analysis using Python courses for student from non-engineering background. I had more that 40 graduate students enrolled from Finance, Accounting and Economics fields at universirty of Alzahra.
Course designer and instructor - University of Tehran
Jan 2022 - Feb 2022
A customized course I designed, featuring numerous finance-related examples using Numpy and Pandas. In this course, I explored the key documentation features of Numpy and Pandas in Python, applying them to solve problems and perform calculations in finance. This course was viewed more than 7K times and had more than 500 enrollments through out years up to now.

Industry Experience

Senior Data Scientist/ML engineer - Iran's National Traffic and Transporation Association
Aug 2022 - Jul 2024
I worked as a data scientists and machine learning pipeline engineer to design fully automated prediction pipelines on traffic flow in critical highways in Tehran metropolitan area. Also, I collaborated on image processing projects and in specific object detection using YOLO technology to detect and count number of cars in desert area parking lots using Drone captured videos.
Research and Development Engineer - Mofid Securities
Sep 2020 - Dec 2021

At Mofid, I was actively collabrotaing in research and development on prediction projects. My responsibilities included

  • Testing neural network architectures to achieve the highest prediction accuracy.
  • Denoising high-frequency data related to stock prices and returns.
  • Backtesting prediction models to evaluate their performance in real-world scenarios.

Education

2025 - present
Ph.D. in Business Analytics
University of Cincinnati
2018 - 2022
Master of Financial Engineering and Risk Management
Ershad Damavand Istitute of Higher Education
2014 - 2018
Bachelor of Financial Management
University of Tehran, Faculty of Management

Projects

Stock Characters.
Asset Pricing Stock Data Python
Stock Characters.
A research project focused on building and analyzing stock-level characteristics for asset pricing applications. The project organizes market and firm-level signals into a reproducible data workflow for studying cross-sectional return patterns and machine learning models in empirical asset pricing.
Traffic Flow Prediction - LSTM Encoder Decoder.
LSTM Encoder-Decoder Python
Traffic Flow Prediction - LSTM Encoder Decoder.
Developed an encoder–decoder architecture based on Long Short-Term Memory (LSTM) networks to forecast short-term traffic flow across three major metropolitan highways in Tehran. The model leverages sequential time-series data to capture complex temporal dependencies and generate multi-step ahead predictions, supporting more efficient traffic management and congestion analysis.
Drone video object detection - YOLO 8.
YOLO 8 Python
Drone video object detection - YOLO 8.
Applied the YOLOv8 framework to drone-captured video data to detect and count vehicles in large-scale parking lots. The system processes aerial footage in real time, leveraging deep learning-based object detection to accurately localize and track cars across frames, enabling scalable parking utilization analysis and monitoring.

Get in Touch

My inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!