Jeffrey Strickland | The George Washington University (original) (raw)
Papers by Jeffrey Strickland
Abstract : PURPOSE: The purpose of this brief to propose an approach for performing "system ... more Abstract : PURPOSE: The purpose of this brief to propose an approach for performing "system of system" level V&V of the BMDS representations. ISSUES: Which interfaces are "truly" SoS level? How do we V&V interfaces? What in DSA corresponds to a Federation Object Model (FOM) in HLA? How do we achieve meta-model validation in the absence of a conceptual model? What constitutes a metadata model?
Interdisciplinary Lively Application Projects
I am often asked if logistic regression is a machine learning algorithm. I say that it is not, fo... more I am often asked if logistic regression is a machine learning algorithm. I say that it is not, for I can formulate it mathematically and solve it using matrix equations, for example. Its solution is derived deterministically and estimation is performed mathematically, through optimization methods. Its link function is a is a mathematical equation. Logistic regression is most appropriate when the dependent variable (target variable) has two possible outcomes (binary). Will customers respond to an offer or unsubscribe, will the enemy fight or flee, will subjects respond to treatment or grow ill, will livestock live or die? Yes or no? One or zero? Here I will take you on a journey into the art and science of predictive modeling using logistic regression, inside-and-out.
Time Series Analysis with Open Source Tools introduces the subject using R and Python programming... more Time Series Analysis with Open Source Tools introduces the subject using R and Python programming and tools. This book assumes a basic understanding of statistics and mathematical or statistical modeling. Although a little programming experience would be nice, it is not required. There are a few “formulas,” with no theorems or proofs, and calculus never appears. Chapters one and two introduce the topic at hand with an overview and a brief discussion about the components of time series. R programming is introduced in Chapter 3 in the R-Studio environment with decomposing and analyzing the components of time series data using unemployment rate and consumer cost index over time as an example. It also introduces differencing and simple smoothing for making sense of the data and demonstrates the analysis of seasonality using beer sales. It introduces dealing with nonstationary time series data using loans as an example. Finally, it covers an alternative time series analysis method using R with airline passenger data. Chapter 4 introduces Python in the iPython environment for manipulating time series data. It covers working with data to format the time series, displaying and plotting the data, examining trend, and smoothing data using meat data from the U.S. Department of Agriculture. It also introduces loading and formatting data that is not native to Python add-ins. Later chapters cover the various application of time series analysis in several different industries including political, financial, and environmental. ARMA, ARIMA, and UCM methods and covered in detail, and GLARMA models are introduced.
Mathematics plays a critical role in many modeling and simulation applications, and whenever deta... more Mathematics plays a critical role in many modeling and simulation applications, and whenever detailed representations of combat phenomenon are required (e.g., attrition, detection, etc.). The required mathematics can seem daunting, but a wealth of problems relevant to the M&S community can be addressed with efficient numerical algorithms and heuristics. We shall present a small sampling of these methods and relevant examples of their usage. Examples include combat attrition, target engagement, detection, and missile defense. Most of these can be implemented with a spreadsheet for demonstration.
Abstract : PURPOSE: The purpose of this brief to propose an approach for performing "system ... more Abstract : PURPOSE: The purpose of this brief to propose an approach for performing "system of system" level V&V of the BMDS representations. ISSUES: Which interfaces are "truly" SoS level? How do we V&V interfaces? What in DSA corresponds to a Federation Object Model (FOM) in HLA? How do we achieve meta-model validation in the absence of a conceptual model? What constitutes a metadata model?
Interdisciplinary Lively Application Projects
I am often asked if logistic regression is a machine learning algorithm. I say that it is not, fo... more I am often asked if logistic regression is a machine learning algorithm. I say that it is not, for I can formulate it mathematically and solve it using matrix equations, for example. Its solution is derived deterministically and estimation is performed mathematically, through optimization methods. Its link function is a is a mathematical equation. Logistic regression is most appropriate when the dependent variable (target variable) has two possible outcomes (binary). Will customers respond to an offer or unsubscribe, will the enemy fight or flee, will subjects respond to treatment or grow ill, will livestock live or die? Yes or no? One or zero? Here I will take you on a journey into the art and science of predictive modeling using logistic regression, inside-and-out.
Time Series Analysis with Open Source Tools introduces the subject using R and Python programming... more Time Series Analysis with Open Source Tools introduces the subject using R and Python programming and tools. This book assumes a basic understanding of statistics and mathematical or statistical modeling. Although a little programming experience would be nice, it is not required. There are a few “formulas,” with no theorems or proofs, and calculus never appears. Chapters one and two introduce the topic at hand with an overview and a brief discussion about the components of time series. R programming is introduced in Chapter 3 in the R-Studio environment with decomposing and analyzing the components of time series data using unemployment rate and consumer cost index over time as an example. It also introduces differencing and simple smoothing for making sense of the data and demonstrates the analysis of seasonality using beer sales. It introduces dealing with nonstationary time series data using loans as an example. Finally, it covers an alternative time series analysis method using R with airline passenger data. Chapter 4 introduces Python in the iPython environment for manipulating time series data. It covers working with data to format the time series, displaying and plotting the data, examining trend, and smoothing data using meat data from the U.S. Department of Agriculture. It also introduces loading and formatting data that is not native to Python add-ins. Later chapters cover the various application of time series analysis in several different industries including political, financial, and environmental. ARMA, ARIMA, and UCM methods and covered in detail, and GLARMA models are introduced.
Mathematics plays a critical role in many modeling and simulation applications, and whenever deta... more Mathematics plays a critical role in many modeling and simulation applications, and whenever detailed representations of combat phenomenon are required (e.g., attrition, detection, etc.). The required mathematics can seem daunting, but a wealth of problems relevant to the M&S community can be addressed with efficient numerical algorithms and heuristics. We shall present a small sampling of these methods and relevant examples of their usage. Examples include combat attrition, target engagement, detection, and missile defense. Most of these can be implemented with a spreadsheet for demonstration.