USD/KZT Forecast - Tenge Exchange Rate Prediction

Light
Dark

Technologies Used

Python: main language for data analysis and model building.

FastAPI: lightweight backend for API delivery.

yfinance: loads historical currency and stock data.

Prophet: Meta’s time series forecasting model (trends, seasonality).

Plotly.js: builds interactive charts on the frontend.

Tailwind CSS: utility-first CSS for modern UI with light/dark modes.

JavaScript: handles form logic, data display, and chart rendering.

Author: @Meruyert Zhaxymurat

Forecast generated using AI. Results may differ from the actual rate.
Date Predicted Lower Upper Trend
Forecast Information

Summary: The forecast uses Prophet — a time-series model from Meta that is part of the Machine Learning (ML) family within the broader Artificial Intelligence (AI) domain.

Below is the logical hierarchy of the model stack.

Visible after forecast generation
AI — Artificial Intelligence
Methods enabling systems to perform tasks requiring human-like intelligence.
→
ML — Machine Learning
A subset of AI: algorithms learn patterns from data.
→
Time Series Forecasting
Predicting future values based on historical trends and seasonality.
→
Prophet
Meta’s library for fast, interpretable time-series forecasts.

Why chosen:

  • Prophet identifies trends and seasonality quickly.
  • External factors (e.g., oil prices) can be added as regressors.
  • Results depend on data quality and external variables.