Errata

Practical Time Series Analysis

Errata for Practical Time Series Analysis

Submit your own errata for this product.

The errata list is a list of errors and their corrections that were found after the product was released.

The following errata were submitted by our customers and have not yet been approved or disproved by the author or editor. They solely represent the opinion of the customer.

Color Key: Serious technical mistake Minor technical mistake Language or formatting error Typo Question Note Update

Version Location Description Submitted by Date submitted
Printed Page p286, 295, 296
in sample codes

I think the book uses perf.write_eval() in the perf library as a system for recording weights and results that are updated at each epoch (in p286, 295, 296 sample code in JapaneseVer).

But, I can't install perf... I looked at PyPI and the perf library doesn't exist... maybe it has been taken out of service or renamed?

I am working with google colab and below is the error text that is displayed as a result of pip install perf.

ERROR: Could not find a version that satisfies the requirement perf (from versions: none)
ERROR: No matching distribution found for perf

 Anonymous  Jan 10, 2023 
Printed Page page 213
5th paragraph

Original:
This sensor is x in the following code, and we plot how the
measured value relates to the actual position in Figure 7-2:

For context, I think the sensor value is z. The variable x is the state value, and the actual observed value is the noise added to the state variable x.

 Anonymous  Aug 17, 2023 
Printed Page 22
footnote

It looks like the reference is missing.

 Ryoko  May 19, 2021 
PDF, Page 33
Both python codes produce an error

On the first python code section we create a aggregate donations by member using the following:

agg_don = donations.groupby('member').apply(lambda df: df.amount.resample("W-MON").sum().dropna())

Immediatly on the following piece of code on the same page, before merging the data, we try to filter the donations using using

member_donations = agg_donations[agg_donations.member
== member]

But agg_don is a series, not a DataFrame, so there's an error:

'Series' object has no attribute 'user'

Here's a notebook for this error:

https://github.com/johnnybigoode/practical-time-series/blob/main/pg33_donations_question.ipynb

 dione bigode  Aug 31, 2021 
Printed Page 57
1st equation

In the first equation it appears a ";" after the first "x". I believe it should not be there.

 Anonymous  May 10, 2022 
Printed Page 93
Last bullet point

Limits for the critical region for the ACF should be +/- 1.96 / sqrt(n), not +/- 1.96 * sqrt(n). This error is repeated on page 95 when reporting limits for the PACF.

 Greg Schwartz  Dec 20, 2021 
ePub Page 183
1st sentence of 1st paragraph

In book, it said "We see significant at lags 3 and 9," but I think significant values are at lag 3 and 10.

Why it is lag 9?

Anonymous  Jul 19, 2021 
Printed Page 183
last paragraph

1. Page 183: the last paragraph, We see significant values at lag 3 and 9, so we fit an MA model with these lags……
a. I believe it’s lag 10 instead of lag 9 based on the ACF plot you showed us

 Fan Yang  Feb 14, 2024 
Printed Page 185
third paragraph

In the inline equation, I believe y_t should be e_t and e_t should be e_{t+1}.

Also, there is a semicolon in the second equation of the page that should not be there.

(By the way, excellent book, I wish all the examples were in Python)

 Javier Cerezo Lafuente  Apr 14, 2022 
Printed Page 185
3rd paragraph

the third paragraph, to predict one time step in the future, out estimate for yt+1¬ is u + theta1 * yt + et
a. I believe the right expression should be theta1 * et + et+1

 Fan Yang  Feb 14, 2024