Back Matter

Search in book...
Toggle Font Controls
Create new playlist

Name your new playlist

Playlist description (optional)
Sign In

Email address

Password

Forgot Password?

or

Continue with Facebook

Continue with Google
Sign Up

Full Name

Email address

Confirm Email Address

Password

or

Continue with Facebook

Continue with Google

Appendix A: Boost Libraries

Introduction

At various points in this book, we need to use the Boost libraries. Specifically, we have two aspects that we need and which are prerequisites for some of the projects. The first is that we use the Boost.Math libraries for some of the basic statistics and also for performing the t-test. Specifically, we use the cdf of the Student's t-distribution to compute the p-value (https://en.wikipedia.org/wiki/Student%27s_t-distribution). The second requirement is that we need a built version of the Boost.Python libraries. This build needs to be consistent with the version of Python that we are targeting in order to build the Python extension modules. In this case, we use Python version 3.8.

This appendix provides a brief guide to installing and building the Boost libraries, specifically addressing the two aspects highlighted earlier. There are several different ways to do this, and this is just one.

The following link provides more detailed instructions if needed: www.boost.org/doc/libs/1_76_0/more/getting_started/windows.html.

Installation

1. Download the Boost libraries and unpack it to install a complete Boost distribution.

The root directory will be oost_1_76_0...

2. Set the system environment variable BOOST_ROOT to the corresponding root directory of the boost installation.

Building

3. We need to build Boost.Python, and so we first need to configure the Boost build.

Open the file oost_1_76_0 oolsuildsrcuser-config.jam

Add the following line:

using python : 3.8 ;

and save the changes.

Open the command prompt and change your current directory to the Boost root directory. Then, type the following command:

bootstrap

The first command prepares the Boost.Build system for use.

Then type .2

The second command invokes Boost.Build to build the separately compiled Boost libraries. Then be patient, as indicated by the Boost.Build system.

Finally, check that the libraries have been built. Specifically, you need the following:

oost_1_76_0stageliblibboost_python38-vc142-mt-gd-x32-1_76.lib
oost_1_76_0stageliblibboost_python38-vc142-mt-x32-1_76.lib
oost_1_76_0stageliblibboost_python38-vc142-mt-gd-x64-1_76.lib
oost_1_76_0stageliblibboost_python38-vc142-mt-x64-1_76.lib

In reality, we do not use the 32 bit libraries in this project; however, it is worth having them available should you want to target Win32.

References

The two main references are

www.boost.org/doc/libs/1_76_0/more/getting_started/windows.html

and

www.boost.org/doc/libs/1_76_0/libs/python/doc/html/building.html

Appendix B: CMAKE

Introduction

This appendix describes the StatsLibCM project. This is a CMake project to build a static library (libStatsLibCM.a) using the MinGW GCC toolset (the same toolset installed by CodeBlocks). The project output (libStatsLibCM.a) should be exactly the same as the outputs produced by building the Debug and Release builds, respectively, with CodeBlocks. The only difference in this project is that we specify everything using CMakeLists.txt.

The advantage of this project is that it can be opened in Visual Studio Community Edition 2019. And the targets can be built from there. Hence, this project could be added to the solution file SoftwareInteroperability.sln , and could substitute usage of CodeBlocks when building the ABI-compatible version of the statistics library for R.

Project Setup

There are two main files in the StatsLibCM project directory. CMakeSettings.json contains the project settings. CMakeLists.txt contains the CMake instructions for building the project. Open CMakeSettings.json (double-click on the file). Figure B-1 shows the main view.

On the left-hand side, we can see the Configurations. There are currently three configurations. The x64-Debug configuration is the Windows default. We have added two further configurations: MinGW64-Debug and MinGW64-Release. When the configurations are added, depending on how your environment is set up, the paths may or may not be correct. The Visual Studio default settings for the MinGW toolset were different from my environment. To correct this requires editing the CMakeSettings.json. Under “Mingw64-Debug”, “environments”, we have added the following block:

{

"MINGW64_ROOT": "D:/Program Files/mingw-w64/x86_64-8.1.0-posix-seh-rt_v6-rev0/mingw64",

"BIN_ROOT": "${env.MINGW64_ROOT}/bin",

"FLAVOR": "x86_64-w64-mingw32",

"TOOLSET_VERSION": "8.1.0",

"PATH":"${env.MINGW64_ROOT}/bin;${env.MINGW64_ROOT}/../usr/local/bin;${env.MINGW64_ROOT}/../usr/bin;${env.MINGW64_ROOT}/../bin;${env.PATH}",

"INCLUDE": "${env.INCLUDE};${env.MINGW64_ROOT}/include/c++/${env.TOOLSET_VERSION};${env.MINGW64_ROOT}/include/c++/${env.TOOLSET_VERSION}/tr1;${env.MINGW64_ROOT}/include/c++/${env.TOOLSET_VERSION}/${env.FLAVOR}",

"environment": "mingw_64"

}

And we have added the same under release. This configures the path to the GCC toolset.

The CMakeLists.txt file contains the CMake configuration. There are a large number of options. The variables are described in the CMake documentation here: https://cmake.org/cmake/help/latest/manual/cmake-variables.7.html.

Building the Outputs

Once the project is opened, select the configuration you want to build and Build All. The targets are output in subdirectories of the StatsLibCM project. You can check that the build output functions correctly in R by editing Makevars.win under the StatsR project, in src. Change the line #PKG_LIBS to point to the debug or the release build of the library:

#PKG_LIBS="<your path>/SoftwareInteroperability/StatsLibCM/out/build/Mingw64-Debug/libStatsLibCM.a

Following this, do a Clean and Rebuild in RStudio and run the tests.

References

Duffy, D.J. & Germani, A. (2013). C# for Financial Markets. John Wiley & Sons, Chichester.

Eddelbuettel D. (2013). Seamless R and C++ Integration with Rcpp. Use R! Springer, New York.

Fletcher, S. & Gardner, C. (2009). Financial Modeling in Python. John Wiley & Sons, Chichester.

Fraser, Stephen R.G. (2009). Pro Visual C++/CLI and the .Net 3.5 Platform. Apress.

Heege, Marcus (2007). Expert Visual C++/CLI .Net for Visual C++ Programmers. Apress.

Kabacoff, Robert. (2015). R in Action. Manning Publications, New York.

Lott, Steven F. (2016). Modern Python Cookbook. Packt Publishing.

Nelli, Fabio. (2018). Python Data Analytics. Apress.

Ragunathan, Vivek (2016). C++/CLI Primer: For .Net Development. Apress.

Wickham, Hadley & Grolemund, Garrett. (2017). R for Data Science. O’Reilly Media, Inc.

Wilkes, Matthew (2020). Advanced Python Development. Apress.

Index

Accord.NET

Application

Console

MFC

Windows GUI

Windows WPF

Application Binary Interface

Architecture

choices

loosely coupled

Arrange-act-assert pattern

Assembly

private

shared

Boost

BOOST_PYTHON_FUNCTION_OVERLOADS macro

BOOST_ROOT

installation

version

C++

classes

codebase

C++17 Standard

high-performance

object oriented

statistical functions

structured bindings

C++/CLI

CMake, StatsLibCM

CmdLet

CodeBlocks

installation

version

Common Language Runtime (CLR) support

Compiler

GCC

gcc toolchain

MSVC

Components

assembly

bridging layer

extension modules

library

package

wrapper

COM wrapper

Correlation coefficient

CPython

Cross-platform development

Data Manager

Datasets

Descriptive statistics

central tendency

kurtosis

maximum

mean

median

minimum

range

skew

spread

standard deviation

variance

VarianceType

enumeration

population

sample

Design pattern

adaptor

facade

pointer-to-implementation (pimpl)

Dynamic-link library (.dll)

Environment

hosting

variable

Excel

Excel-DNA

workbook

Frameworks

Boost.Python

C++/CLI

CPython

PyBind

Rcpp

Function

BOOST_PYTHON_FUNCTION_OVERLOADS macro

f_overloads structure

overloaded

Garbage Collector (GC)

gdb

Global Assembly Cache (GAC)

GNU Compiler Collection (GCC)

gnuplot

GoogleTest

H, I

Header-only library

Hypothesis test

J, K

Jupyter

Library

dynamic-link library (.dll)

static library (.lib)

Linear regression

house prices

modelling

Machine learning (ML)

Makefile

Makeconf

Makevars.win

Managed runtime

Managed wrapper

Marshalling

Memory management

destructor and finalizer

managed memory

unmanaged memory

Metadata

Microsoft Intermediate Language (MSIL)

MinGW

CodeBlocks

Rtools

“Moving Average” menu

N, O

NuGet Package Manager

Performance

bench:mark

PerformanceTest.py

pimpl design pattern

Pinning

PowerShell

Projects

StatsClient

StatsCLR

StatsCLR.UnitTests

StatsConsole

StatsDll

StatsExcel

StatsLib

StatsLibCB

StatsLibTest

StatsPython

StatsPythonBoost

StatsPythonPyBind

StatsPythonRaw

StatsR

StatsViewer

Project types

PyArg_ParseTuple template string

PyBind

installation

return value policies

Python

PyObject

REPL

shell

version

QuantLib

environment

language

version

Rcpp

framework

installation

modules

namespace

RCPP_MODULE macro

version

Reflection

Resource Acquisition is Initialization (RAII)

Rgui

RObject

RStudio Desktop

IDE

installation

version

Rtools

installation

version

Shiny App, Distribution Explorer

Single Document Interface (SDI)

Software ecosystem

Software Interoperability

project

solution file, SoftwareInteroperability.sln

Standard Library

Static library (.lib)

Statistical tests

FTest class

StatisticalTest class

TTest class

z-test

ZTest class

Statistics

descriptive statistics

functions

statistical_functions, map

Statistics service

Testing

arrange-act-assert pattern

GoogleTest

MSUnitTest

testthat

Time series, moving average

Toolchain

Tools, cross-platform

Toolset

T-Test (Student’s)

from data

one-sample

two-sample

Types

built-in

STL

user-defined

DataManager

FTest

TTest

Unicode

UTF-8

UTF-16

VARIANT

Visual Studio Code

Visual Studio Community Edition 2019

Windows

application

platform

X, Y

x64 builds

x86/Win32 target

Z-Test

from data

one-sample

two-sample

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

Table of Contents for Back Matter

Create new playlist

Sign In

Sign Up

Table of Contents for
Back Matter