Select A Store
Select Store
My Store:
Select A Store
Time Series Forecasting Predicting Trends with ML: Master ML for finance, weather, and demand forecasting
by Myles, Isandro
Availability:
-
Online Only
$15.90
Overview
Overview
When data has a clock, everything changes. Time Series Forecasting: Predicting Trends with ML shows you how to turn noisy historical data into reliable, decision-ready forecasts-for finance, weather, energy, retail demand, traffic, sensors, and more. No mystique, no hand-waving-just the tools, patterns, and trade-offs that actually work in production.
You'll learn the full playbook: build solid baselines, add seasonality and holiday effects, bring in external drivers, and graduate to advanced ML and deep learning when it truly adds lift. You'll backtest honestly, quantify uncertainty, and communicate results so stakeholders can act with confidence.
What you'll be able to do:
Frame forecasting problems that match business goals (point, probabilistic, and scenario forecasts)
Build and compare ARIMA/SARIMA, ETS, Prophet-style additive models, and state-space approaches
Engineer time-aware features (lags, windows, Fourier terms, holidays, weather, prices)
Train ML models for sequences (gradient boosting, TFT/Transformers) and avoid leakage traps
Evaluate with rolling backtests; report MAPE/WAPE/SMAPE and calibration the right way
Reconcile hierarchies, handle intermittent demand, detect changepoints, and stress-test robustness
Operate at scale with automated pipelines, monitoring, and champion/challenger deployment
Written for analysts, engineers, quants, and builders, this is your practical guide to forecasts that are defendable, explainable, and useful on day one.
- Format: Trade Paperback
- Author: Myles, Isandro
- ISBN: 9798265007537
- Condition: New
- Dimensions: 9.00 x 0.35
- Number Of Pages: 162
- Publication Year: 2025
Language:
English