Frank L.

Stat packages, data analyst, tutor

Stat packages, data analyst, tutor

GIS

I taught STATA thematic mapping, daylong seminars. I used ArcGIS from the step of importing "important datasets" one medical, the other political and merging along with STATA, SAS or SPSS and creating a thematic map showing a correlation. Presently, I am dabbling in R mapping (again involved with importing data and shapefiles).

MATLAB

Over the last 30 years I worked as a statistical programmer/consultant at NYU where I supported Matlab. My degrees are in math and I taught introductory tutorials in Matlab. I'm comfortable with matrix algebra and can usually work out problems. Tell me what you're trying to do in Matlab and I'll figure out how/if I can help you.

R

Along with STATA, ArcGIS, I've used R as a cartographer and additionally for statistical analysis. I subscribe to CRAN, the R archive network to keep current. I've written an article comparing R to other stat packages. I've taught tutorials at NYU introducing R. Since R is free it will always be high on my list of packages.

SAS

Taught SAS Intro and SAS Intermediate for 15 years at NYU Data Services. Consulted in using SAS as a front-end for MySQL. Wrote NYU/Connect article SAS for SPSS programmers. Proficient in SAS Proc GMap to make thematic maps. Worked with SAS callable Sudaan. SAS like all the statistical packages has three basic skill sets which I can teach. One is using SAS to create data sets, using Proc Format to develop good meta-data to document the data. Merging, subsetting and other tools to develop good datasets. Also at this level, one must become involved with weights for complex data. Secondly, using elementary statistical procedures to clean the data. Also important is using procedures to visualize the data. Third, one must use advanced statistical tests such as regression to investigate the hypothesis of the research.

SPSS

SPSS was Frank's first stat package in the '80s at Cornell where he was the Stat/data manager for a large EPA project studying dirnking water. While tutoring and consulting students and researchers in SPSS SAS, Stata and R at NYU Data Services for many years, he also consulted for NYC Children's Services and Lancome. SPSS is the most user-friendly stat package. It incorporates two complete tool boxes. The first allows us to manage data and to describe data and the second allows us to perform statistical analysis. Data management: handling large data files, formatting them to be useful while doing analysis and manipulating them to add new variables derived from the original data and to clean data of mistakes is 1/2 of the work we SPSS users perform. Analysis and report writing are often the work which follows. After years of working with every step of data work - from creating questionnaires to inputting the data then to cleaning, describing and analyzing, my experience often makes me a useful asset to the student/researcher at most levels.

STATA

I have supported Stata (along with other stat packages) at NYU Data Services Group for 20 years. A former actuary, I moved to acad. research at Cornell by participating in a large EPA grant as stat/dba administrator. I have consulted with NYC Children's Services, and taught Stata for Timberlakes Consultancy (an Oxford U affiliated group). Presently, I'm working with Stata/Mapping visualization tools for thematic mapping, the set of commands around "spmap". Stata may be approached from several directions: 1) as a data management package, 2) as a statistical analysis tool, and 3) as a data visualization tool. Learning data management involves a knowledge of data types, data attributes and formats. It also involves importing data, reshaping and merging data files, logical expressions, operators, operator precedence and functions. Working with missing values in very important. An overview of statistical tests involves tests involving one variable, and tests involving several variables (to look at relationships of several variables). Visualization of data involves graphics including thematic mapping.

SPSS

SPSS was Frank's first stat package in the '80s at Cornell where he was the Stat/data manager for a large EPA project studying dirnking water. While tutoring and consulting students and researchers in SPSS SAS, Stata and R at NYU Data Services for many years, he also consulted for NYC Children's Services and Lancome. SPSS is the most user-friendly stat package. It incorporates two complete tool boxes. The first allows us to manage data and to describe data and the second allows us to perform statistical analysis. Data management: handling large data files, formatting them to be useful while doing analysis and manipulating them to add new variables derived from the original data and to clean data of mistakes is 1/2 of the work we SPSS users perform. Analysis and report writing are often the work which follows. After years of working with every step of data work - from creating questionnaires to inputting the data then to cleaning, describing and analyzing, my experience often makes me a useful asset to the student/researcher at most levels.

MATLAB

Over the last 30 years I worked as a statistical programmer/consultant at NYU where I supported Matlab. My degrees are in math and I taught introductory tutorials in Matlab. I'm comfortable with matrix algebra and can usually work out problems. Tell me what you're trying to do in Matlab and I'll figure out how/if I can help you.

R

Along with STATA, ArcGIS, I've used R as a cartographer and additionally for statistical analysis. I subscribe to CRAN, the R archive network to keep current. I've written an article comparing R to other stat packages. I've taught tutorials at NYU introducing R. Since R is free it will always be high on my list of packages.

SPSS

SPSS was Frank's first stat package in the '80s at Cornell where he was the Stat/data manager for a large EPA project studying dirnking water. While tutoring and consulting students and researchers in SPSS SAS, Stata and R at NYU Data Services for many years, he also consulted for NYC Children's Services and Lancome. SPSS is the most user-friendly stat package. It incorporates two complete tool boxes. The first allows us to manage data and to describe data and the second allows us to perform statistical analysis. Data management: handling large data files, formatting them to be useful while doing analysis and manipulating them to add new variables derived from the original data and to clean data of mistakes is 1/2 of the work we SPSS users perform. Analysis and report writing are often the work which follows. After years of working with every step of data work - from creating questionnaires to inputting the data then to cleaning, describing and analyzing, my experience often makes me a useful asset to the student/researcher at most levels.

STATA

I have supported Stata (along with other stat packages) at NYU Data Services Group for 20 years. A former actuary, I moved to acad. research at Cornell by participating in a large EPA grant as stat/dba administrator. I have consulted with NYC Children's Services, and taught Stata for Timberlakes Consultancy (an Oxford U affiliated group). Presently, I'm working with Stata/Mapping visualization tools for thematic mapping, the set of commands around "spmap". Stata may be approached from several directions: 1) as a data management package, 2) as a statistical analysis tool, and 3) as a data visualization tool. Learning data management involves a knowledge of data types, data attributes and formats. It also involves importing data, reshaping and merging data files, logical expressions, operators, operator precedence and functions. Working with missing values in very important. An overview of statistical tests involves tests involving one variable, and tests involving several variables (to look at relationships of several variables). Visualization of data involves graphics including thematic mapping.

Epidemiology

My 30 year tenure at NYU included faculty status at NYU Med School. I consulted and did Public Health analysis using Medicare, Ambulatory Care (ER) , World Health data. My last f/t job was at Cornell Med where I worked with Stata, SAS, SPSS comparing British and US health delivery. I worked on a project defining clusters of good and bad MDs. My first data analysis position (after being a math prof) was on an EPA water quality grant. I've done survival analysis and am now working with multilevel modeling. I do thematic mapping and spatial analysis which is useful in PH projects.

Epidemiology

My 30 year tenure at NYU included faculty status at NYU Med School. I consulted and did Public Health analysis using Medicare, Ambulatory Care (ER) , World Health data. My last f/t job was at Cornell Med where I worked with Stata, SAS, SPSS comparing British and US health delivery. I worked on a project defining clusters of good and bad MDs. My first data analysis position (after being a math prof) was on an EPA water quality grant. I've done survival analysis and am now working with multilevel modeling. I do thematic mapping and spatial analysis which is useful in PH projects.