Trident Help

RPN Power 48 manual · Chapter 13 of 20

Statistics

Collect data in ΣDAT, then mean, standard deviation, totals, line and curve fits, predictions and correlation from the STAT menu.

13.1Enter data with Σ+#

Statistics work on ΣDAT, a matrix variable with one row per data point. Σ+ adds the number on level 1 as a new row; a vector adds one row with several columns (x and y), and a matrix adds all its rows at once. Σ- removes the last row and puts it back on the stack, NΣ counts the rows and CLΣ deletes ΣDAT. The STAT menu is on the third page of APPS.

STAT page 1: Σ+, Σ-, CLΣ, NΣ, TOT, MEAN. The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1: /   Σ+   Σ-   CLΣ   NΣ   TOT  MEAN .
STAT page 1: Σ+, Σ-, CLΣ, NΣ, TOT, MEAN.
STAT page 2: SDEV, VAR, MAXΣ, MINΣ, PSDEV, PVAR. The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1: /  SDEV  VAR  MAXΣ MINΣ PSDEV PVAR .
STAT page 2: SDEV, VAR, MAXΣ, MINΣ, PSDEV, PVAR.
STAT page 3: XCOL, YCOL, LR, PREDY, PREDX, CORR. The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1: /  XCOL YCOL   LR  PREDYPREDX CORR .
STAT page 3: XCOL, YCOL, LR, PREDY, PREDX, CORR.
STAT page 4: COV, PCOV and the four fit models. The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1: /  COV  PCOV LINFI LOGFIEXPFI PWRFI.
STAT page 4: COV, PCOV and the four fit models.
STAT page 5: BESTFIT and the sums ΣX, ΣY, ΣX2, ΣY2, ΣXY. The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1: / BESTF  ΣX    ΣY   ΣX2  ΣY2   ΣXY .
STAT page 5: BESTFIT and the sums ΣX, ΣY, ΣX2, ΣY2, ΣXY.
Example 13.1

Enter three numbers and count them

  1. Open STAT, then type each number and press Σ+ (first key). NΣ is the fourth key.

    APPSNXTNXTF12F14F19F1F4
Result of the example "Enter three numbers and count them" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:                              3 /   Σ+   Σ-   CLΣ   NΣ   TOT  MEAN .
You should see1: 3
Example 13.2

Take the last point back out

  1. Σ- (second key) removes 9 from ΣDAT and returns it.

    APPSNXTNXTF12F14F19F1F2
Result of the example "Take the last point back out" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:                              9 /   Σ+   Σ-   CLΣ   NΣ   TOT  MEAN .
You should see1: 9

13.2Mean, standard deviation and totals#

MEAN, TOT (the sum), MAXΣ and MINΣ work column by column: with one column the answer is a number, with several it is a vector. SDEV and VAR are the sample standard deviation and variance (dividing by n − 1); PSDEV and PVAR are the population versions (dividing by n).

xˉ=1n∑xi,s=∑(xi−xˉ)2n−1\bar{x}=\frac{1}{n}\sum x_i,\qquad s=\sqrt{\frac{\sum (x_i-\bar{x})^{2}}{n-1}}
Example 13.3

Mean of 2, 4 and 9

  1. Enter the data, then MEAN (sixth key).

    APPSNXTNXTF12F14F19F1F6
Result of the example "Mean of 2, 4 and 9" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:                              5 /   Σ+   Σ-   CLΣ   NΣ   TOT  MEAN .
You should see1: 5
Example 13.4

Sample standard deviation

  1. SDEV is the first key on page 2.

    APPSNXTNXTF12F14F19F1NXTF1
Result of the example "Sample standard deviation" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:                  3.60555127546 /  SDEV  VAR  MAXΣ MINΣ PSDEV PVAR .
You should see1: 3.60555127546
Example 13.5

Population standard deviation

  1. PSDEV is the fifth key on page 2.

    APPSNXTNXTF12F14F19F1NXTF5
Result of the example "Population standard deviation" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:                  2.94392028878 /  SDEV  VAR  MAXΣ MINΣ PSDEV PVAR .
You should see1: 2.94392028878
Example 13.6

Means of two columns

  1. Enter the points (1, 2), (2, 4), (3, 7) as a matrix, Σ+, then MEAN: one mean per column.

    ◀[ ]9◀[ ]91SPC2▶▶◀[ ]92SPC4▶▶◀[ ]93SPC7ENTERAPPSNXTNXTF1F1F6
Result of the example "Means of two columns" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:            [ 2 4.33333333333 ] /   Σ+   Σ-   CLΣ   NΣ   TOT  MEAN .
You should see[ 2 4.33333333333 ]

13.3Fit a line or curve and predict#

With two columns, LR fits y = b + m·x by least squares and returns the intercept and slope, tagged. They are also kept in ΣPAR, so PREDY gives the y for an x and PREDX the x for a y. CORR is the correlation coefficient, COV and PCOV the sample and population covariance. XCOL and YCOL choose which columns are x and y (1 and 2 at first).

m=∑(xi−xˉ)(yi−yˉ)∑(xi−xˉ)2,b=yˉ−mxˉm=\frac{\sum (x_i-\bar{x})(y_i-\bar{y})}{\sum (x_i-\bar{x})^{2}},\qquad b=\bar{y}-m\bar{x}
Example 13.7

Fit a line

  1. Enter (1, 2), (2, 4), (3, 7), Σ+, then LR (page 3, third key).

    ◀[ ]9◀[ ]91SPC2▶▶◀[ ]92SPC4▶▶◀[ ]93SPC7ENTERAPPSNXTNXTF1F1NXTNXTF3
Result of the example "Fit a line" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2:      Intercept: -.666666666667 / 1:                     Slope: 2.5 /  XCOL YCOL   LR  PREDYPREDX CORR .
You should seeIntercept: -.666666666667 Slope: 2.5
Example 13.8

Predict y at x = 4

  1. After LR, drop the two results, type 4 and press PREDY (fourth key).

    ◀[ ]9◀[ ]91SPC2▶▶◀[ ]92SPC4▶▶◀[ ]93SPC7ENTERAPPSNXTNXTF1F1NXTNXTF3←←4F4
Result of the example "Predict y at x = 4" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:                  9.33333333333 /  XCOL YCOL   LR  PREDYPREDX CORR .
You should see1: 9.33333333333
Example 13.9

Correlation

  1. CORR is the sixth key on page 3.

    ◀[ ]9◀[ ]91SPC2▶▶◀[ ]92SPC4▶▶◀[ ]93SPC7ENTERAPPSNXTNXTF1F1NXTNXTF6
Result of the example "Correlation" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:                  .993399267799 /  XCOL YCOL   LR  PREDYPREDX CORR .
You should see1: .993399267799
Model (STAT page 4)FitsFitted as
LINFITy = b + m·xy against x
LOGFITy = b + m·ln xy against ln x
EXPFITy = b·e^(m·x)ln y against x
PWRFITy = b·x^mln y against ln x

Choose a model before LR. BESTFIT (page 5, first key) tries all four, keeps the one with the largest correlation (models that would need the logarithm of zero or a negative number are skipped) and stores its intercept and slope in ΣPAR, ready for PREDY.

Example 13.10

Let BESTFIT pick a power curve

  1. Enter (1, 3), (2, 12), (3, 27), which lie on y = 3x², then Σ+ and BESTFIT.

    ◀[ ]9◀[ ]91SPC3▶▶◀[ ]92SPC12▶▶◀[ ]93SPC27ENTERAPPSNXTNXTF1F1NXTNXTNXTNXTF1
  2. Predict y at x = 4 with PREDY (page 3, fourth key).

    4NXTNXTNXTF4
Result of the example "Let BESTFIT pick a power curve" The display shows: RAD                      { HOME } / 6: / 5: / 4: / 3: / 2: / 1:                             48 /  XCOL YCOL   LR  PREDYPREDX CORR .
You should see1: 48

HP 48 and HP 50g is a trademark of HP Inc.. Trident Calculator is an independent product by South View Studios and is not affiliated with, sponsored by or endorsed by HP. The name is used only to describe the kind of calculator this model resembles; Trident's names, keys, display, fonts and software are its own clean-room design.