Process Help: Correlation and Regression
TiWebly
New Altair Community Member
Hello!
I have the following data (74 observations) and am trying to identify the process and operators to conduct an exploratory analysis (Correlation and Regression) of the relationship between the president's political party affiliation and this consumer good's Producer Price Index value which is averaged over each year annually. I also have the PPI values quarterly and monthly for the same years which will be assessed using the same process in hopes of a more statically significant value.
- Additionally, how to identify which President had the largest change+/- in PPI over their Term and which year (assuming I'm using monthly values for PPI)?
| Years (after inauguration) | President | Party | PPI Observation Date | PPI_YrAvg |
| 1947 | Harry S. Truman | Democrat | 1/1/1947 | 22.5 |
| 1948 | Harry S. Truman | Democrat | 1/1/1948 | 25.5 |
| 1949 | Harry S. Truman | Democrat | 1/1/1949 | 27.4 |
| 1950 | Harry S. Truman | Democrat | 1/1/1950 | 28.4 |
| 1951 | Harry S. Truman | Democrat | 1/1/1951 | 33.2 |
| 1952 | Harry S. Truman | Democrat | 1/1/1952 | 32.1 |
| 1953 | Dwight D. Eisenhower | Republican | 1/1/1953 | 31.9 |
| 1954 | Dwight D. Eisenhower | Republican | 1/1/1954 | 31.9 |
| 1955 | Dwight D. Eisenhower | Republican | 1/1/1955 | 33.6 |
| 1956 | Dwight D. Eisenhower | Republican | 1/1/1956 | 35.7 |
| 1957 | Dwight D. Eisenhower | Republican | 1/1/1957 | 36.5 |
| 1958 | Dwight D. Eisenhower | Republican | 1/1/1958 | 36.5 |
| 1959 | Dwight D. Eisenhower | Republican | 1/1/1959 | 35.7 |
| 1960 | Dwight D. Eisenhower | Republican | 1/1/1960 | 35.5 |
| 1961 | John F. Kennedy | Democrat | 1/1/1961 | 37.5 |
| 1962 | John F. Kennedy | Democrat | 1/1/1962 | 37.5 |
| 1963 | Lyndon B. Johnson | Democrat | 1/1/1963 | 37.6 |
| 1964 | Lyndon B. Johnson | Democrat | 1/1/1964 | 37.9 |
| 1965 | Lyndon B. Johnson | Democrat | 1/1/1965 | 39.5 |
| 1966 | Lyndon B. Johnson | Democrat | 1/1/1966 | 39.5 |
| 1967 | Lyndon B. Johnson | Democrat | 1/1/1967 | 39.4 |
| 1968 | Lyndon B. Johnson | Democrat | 1/1/1968 | 40.9 |
| 1969 | Richard M. Nixon | Republican | 1/1/1969 | 42.2 |
| 1970 | Richard M. Nixon | Republican | 1/1/1970 | 46.2 |
| 1971 | Richard M. Nixon | Republican | 1/1/1971 | 47.4 |
| 1972 | Richard M. Nixon | Republican | 1/1/1972 | 48.4 |
| 1973 | Richard M. Nixon | Republican | 1/1/1973 | 49.2 |
| 1974 | Gerald Ford | Republican | 1/1/1974 | 53.2 |
| 1975 | Gerald Ford | Republican | 1/1/1975 | 59.7 |
| 1976 | Gerald Ford | Republican | 1/1/1976 | 62.3 |
| 1977 | Jimmy Carter | Democrat | 1/1/1977 | 67.5 |
| 1978 | Jimmy Carter | Democrat | 1/1/1978 | 72.8 |
| 1979 | Jimmy Carter | Democrat | 1/1/1979 | 80.5 |
| 1980 | Jimmy Carter | Democrat | 1/1/1980 | 88.7 |
| 1981 | Ronald Reagan | Republican | 1/1/1981 | 96.9 |
| 1982 | Ronald Reagan | Republican | 1/1/1982 | 100.0 |
| 1983 | Ronald Reagan | Republican | 1/1/1983 | 110.0 |
| 1984 | Ronald Reagan | Republican | 1/1/1984 | 115.9 |
| 1985 | Ronald Reagan | Republican | 1/1/1985 | 123.5 |
| 1986 | Ronald Reagan | Republican | 1/1/1986 | 126.3 |
| 1987 | Ronald Reagan | Republican | 1/1/1987 | 125.0 |
| 1988 | Ronald Reagan | Republican | 1/1/1988 | 130.2 |
| 1989 | George Bush | Republican | 1/1/1989 | 136.4 |
| 1990 | George Bush | Republican | 1/1/1990 | 133.4 |
| 1991 | George Bush | Republican | 1/1/1991 | 138.9 |
| 1992 | George Bush | Republican | 1/1/1992 | 138.3 |
| 1993 | Bill Clinton | Democrat | 1/1/1993 | 139.5 |
| 1994 | Bill Clinton | Democrat | 1/1/1994 | 140.3 |
| 1995 | Bill Clinton | Democrat | 1/1/1995 | 144.3 |
| 1996 | Bill Clinton | Democrat | 1/1/1996 | 143.1 |
| 1997 | Bill Clinton | Democrat | 1/1/1997 | 142.8 |
| 1998 | Bill Clinton | Democrat | 1/1/1998 | 144.1 |
| 1999 | Bill Clinton | Democrat | 1/1/1999 | 144.2 |
| 2000 | Bill Clinton | Democrat | 1/1/2000 | 144.9 |
| 2001 | George W. Bush | Republican | 1/1/2001 | 143.9 |
| 2002 | George W. Bush | Republican | 1/1/2002 | 144.3 |
| 2003 | George W. Bush | Republican | 1/1/2003 | 145.2 |
| 2004 | George W. Bush | Republican | 1/1/2004 | 148.9 |
| 2005 | George W. Bush | Republican | 1/1/2005 | 162.5 |
| 2006 | George W. Bush | Republican | 1/1/2006 | 171.5 |
| 2007 | George W. Bush | Republican | 1/1/2007 | 198.7 |
| 2008 | George W. Bush | Republican | 1/1/2008 | 250.3 |
| 2009 | Barack Obama | Democrat | 1/1/2009 | 248.9 |
| 2010 | Barack Obama | Democrat | 1/1/2010 | 261.6 |
| 2011 | Barack Obama | Democrat | 1/1/2011 | 297.9 |
| 2012 | Barack Obama | Democrat | 1/1/2012 | 304.5 |
| 2013 | Barack Obama | Democrat | 1/1/2013 | 315.4 |
| 2014 | Barack Obama | Democrat | 1/1/2014 | 323.4 |
| 2015 | Barack Obama | Democrat | 1/1/2015 | 338.6 |
| 2016 | Barack Obama | Democrat | 1/1/2016 | 345.5 |
| 2017 | Donald Trump | Republican | 1/1/2017 | 341.6 |
| 2018 | Donald Trump | Republican | 1/1/2018 | 358.3 |
| 2019 | Donald Trump | Republican | 1/1/2019 | 362.2 |
| 2020 | Donald Trump | Republican | 1/1/2020 | 369.9 |
This is as far as I've got without guidance.
Please be kind, I'm just starting to explore rapidminer in school and I'd really appreciate any help or advice, Thanks!
0
Answers
-
Hi!
I would first calculate the yearly change of PPI. I assume the formula for that would be PPI_YrAvg / previous(PPI_YrAvg) - 1. You can do this in Excel or in RapidMiner with Differentiate or Lag and Generate Attributes.
Here's an introduction to Generate Attributes: https://academy.rapidminer.com/learn/video/generate-attributes
And two videos for time series calculations: https://academy.rapidminer.com/catalog?query=differentiate
When you have the new column YearlyChange, you can easily use Aggregate to group by party and calculate the average, min and max yearly change.
Regards,
Balázs
1
