Errata

I think that a function name is missing, so the line should read instead:
"double t1 = act365_.year_fraction(settle_date(), d1);"
I would also think that a similar expression for t2 should also use act365_ (instead of act_365()) - for more consistency.

Note from the Author or Editor:
#2
Good catch, thanks very much.

The code line:
`double t1 = act365_(settle_date(), d1);`

should actually be:
`double t1 = act_365().year_fraction(settle_date(), date_01);`

as I removed the functor per recommendation from a technical reviewer (more expressive when it spells out what it does), so the error in the book would fail to compile.

2025.06.24: In dh_errata_july_2025 branch: Changed code in YieldCurve::discount_factor(.) example from

`double t1 = act365_(settle_date(), d1);`

to

`double t1 = act_365().year_fraction(settle_date(), date_01);`

as the functor was removed in favor of the year_fraction(.) member function.

**Note: Not complete per pdf -- replace date_01 with d1**

#14

The line says:

"\frac{2}{3} \lt \frac{4}{6}, because 4 > 2"

This line should be replaced, because in the code, the fraction is simplified at construction.

Instead, replace it with

"\frac{2}{3} \lt \frac{3}{10}, because 3 > 2"

Of course, do not include the quotation marks; I used them here to indicate what should be removed and what should be used to replace the 1st line.

I found it in the print version, but it is also in the other formats.


2025.06.24: Updated in dh_errata_july_2025...

Edit ch02.asciidoc: Line 1104 changed to:

$\frac{2}{3} 2

Now comparing 2/3 and 3/10 with lexicographic comparison (which shows this method fails, as desired).

Note: Verified fixed in pdf generated from branch.

Where is integ_scens defined? (is it supposed to be corr_scens?)

Note from the Author or Editor:
#12

Confirmed, but this is actually on p 310 in Ch 8.

The code block:

```
for (unsigned j = 1; j < integ_scens.cols(); ++j)
{
for (unsigned i = 0; i < integ_scens.rows(); ++i)
{
integ_scens(i, j) = gen_price(integ_scens(i, j - 1), cov_basket(i, i),
corr_norms(i, j - 1));
}
}

```
should be replaced by:

```
for (unsigned j = 1; j < corr_scens.cols(); ++j)
{
for (unsigned i = 0; i < corr_scens.rows(); ++i)
{
corr_scens(i, j) = gen_price(corr_scens(i, j - 1),
cov_basket(i, i), corr_scens(i, j));
}
}

```

2025.06.24: Corrected in dh_errata_july_2025 branch:
Edit ch08.asciidoc: Code in lines 1887-1814 has been changed to:

for (unsigned j = 1; j < corr_scens.cols(); ++j)
{
for (unsigned i = 0; i < corr_scens.rows(); ++i)
{
corr_scens(i, j) = gen_price(corr_scens(i, j - 1),
cov_basket(i, i), corr_scens(i, j));
}
}

The corr_scens(.) previously used an incorrect function name, integ_scens(.).

**Note: The code has been fixed, but the spacing needs to be adjusted (replace tabs with 4 spaces)** Follow-up: This has been fixed and checked.

I think that constructor meant to be a move constructor (and not a copy constructor). So && instead of &.

Note from the Author or Editor:
#4

The sample code line
`OptionInfo(OptionInfo& rhs) = delete;`

requires an additional ampersand:
`OptionInfo(OptionInfo&& rhs) = delete;`

as it is a move constructor.

As an aside, these lines were not in the sample code and should be included (author's responsibility -- on GitHub)

2025.06.24: Corrected in dh_errata_july_2025:

Edit ch03.asciidoc:
The move constructor

`OptionInfo(OptionInfo& rhs) = delete;`

requires an additional ampersand

`OptionInfo(OptionInfo&& rhs) = delete;`

This change has been made.

Note: verified in pdf generated from branch

I think that the norm_pdf lambda (commented as "Standard Normal pdf") does not follow the definition of N'(x) earlier on the same page.

Note from the Author or Editor:
#6

This is incorrect in the book but correct in the sample code on GitHub. The lambda function at the bottom of the page should read:

```
// N'(x): Standard Normal pdf:
auto norm_pdf = [](double x) -> double
{
using namespace std::numbers;
return (inv_sqrtpi / sqrt2) * exp(-x * x/2.0);
};

```

Corrected in dh_errata_july_2025:
Edit ch04.asciidoc: Standard Normal pdf lambda now correctly reads:

// N'(x): Standard Normal pdf:
auto norm_pdf = [](double x)
{
using namespace std::numbers;
return (inv_sqrtpi / sqrt2) * exp(-x * x/2.0);
};

The inv_sqrtpi and the square of x were missing (was only one x). This was correct in the sample code but not in the book.

**Note: spacing in updated pdf is not correct -- fix this.** Otherwise OK.

auto corp = corp = PayoffType::Put;

probably does not need corp twice.

Note from the Author or Editor:
#7

`auto corp = corp = PayoffType::Put;`

should be changed to

`auto corp = PayoffType::Put;`

Revised in dh_errata_july_2025 branch:

Edit ch04.asciidoc, Line 1370:

Changed (the error in)

`auto corp = corp = PayoffType::Put;`

to

`auto corp = PayoffType::Put;`

Note: Verified this is fixed in pdf generated from branch.