Top critical review
2.0 out of 5 starsDoes not actually show you how to forecast.
Reviewed in the United States on January 29, 2023
This book explains the concepts in a non-mathematical way. So, if you do not have a mathematical background then you should be able to follow along. Codes in the book sometimes do not work but if you look at the github page for the book, these errors have been corrected.
My biggest problem with the book is that it does not show you how to forecast beyond your dataset. For example, if you have a monthly/weekly dataset ranging from 2010-2022, then this book does not show you how to forecast into the year 2023. All the models (at least the classic models that the book has talked about, I haven't looked at the deep learning ones yet) divide the dataset into training and test data and then show you how the model is performing against the test data. Now, that we know the models are performing well, how about forecasting beyond the test dataset? That part is missing in the book. So, in my opinion the book helps you to understand forecast theory in a non-mathematical way but does not actually show you how to forecast past the test data that you have available. Another negative is that the book repeats a lot of stuff. It seems like the author is justifying the high price of the book with the number of pages (at close to sixty dollars, this book is pretty expensive for what it is trying to teach). So, two stars for the theoretical part and took away three stars for not showing how to forecast past the test dataset.
When someone buys a book on forecast, they normally hope to learn how to forecast into the future and not just learn how forecasting works within their available dataset. This book fails miserably in this particular direction.