From 0c1d68ee8cc2a631d6029285e771ebbfe119995d Mon Sep 17 00:00:00 2001
From: wukong <wukong@longaeva>
Date: Sat, 7 Dec 2019 15:02:59 -0800
Subject: code formatting changes to lin_reg scripts; added range to output of
 mean scripts;

---
 ABOUT.TXT    |  2 +-
 diff.awk     |  2 +-
 diff1.awk    | 30 +++++++++++++++++-------------
 lin_reg1.awk |  1 -
 lin_reg2.awk |  1 -
 mean.awk     | 17 +++++++++--------
 mean_avg.awk |  9 +++++----
 7 files changed, 33 insertions(+), 29 deletions(-)

diff --git a/ABOUT.TXT b/ABOUT.TXT
index b16ad44..07ab7ec 100644
--- a/ABOUT.TXT
+++ b/ABOUT.TXT
@@ -1,2 +1,2 @@
 repo:	awk;
-desc:	experiments in awk, etc.;
+desc:	experiments in using awk for mathematical and statistical calculations;
diff --git a/diff.awk b/diff.awk
index ba452c7..bf8b472 100644
--- a/diff.awk
+++ b/diff.awk
@@ -1,7 +1,7 @@
 #!/usr/bin/awk -f
 
 ### diff.awk
-# numerical diff along columns
+# print numerical diff along columns
 
 BEGIN {
     OFS = FS
diff --git a/diff1.awk b/diff1.awk
index cf1f2ff..7322e98 100644
--- a/diff1.awk
+++ b/diff1.awk
@@ -1,9 +1,11 @@
 #!/usr/bin/awk -f
 
 ### diff1.awk
-# numerical diff along columns
+# print numerical diff along columns
 
 BEGIN {
+    OFS = FS
+    # OFMT = "%.9g"
     sign = "[+-]?"
     decimal = "[0-9]+[.]?[0-9]*"
     fraction = "[.][0-9]*"
@@ -12,7 +14,7 @@ BEGIN {
 }
 
 NR == 1 {
-    ### orig data columns
+    # orig data columns
     for (n=1; n<=NF; n++) {
         if ($n !~ number) {
             header[n] = $n
@@ -31,18 +33,20 @@ NF {
     if (NF > nf_max)
         nf_max = NF
 
-    ### data columns
-    for (y=1; y<=nf_max; y++) {
-        if ($y == header[y] || $y == dheader[y])
+    # data columns
+    for (n=1; n<=nf_max; n++) {
+        if ($n == header[n] || $n == dheader[n])
             continue
-        if ($y ~ number) {
-            data[y] = $y
-            (data_prev[y] ~ number) ? diff[y] = data[y] - data_prev[y] : diff[y] = ""
-            data_prev[y] = data[y]
-            printf(OFMT, diff[y])
+        if ($n ~ number) {
+            data[n] = $n
+            if (data_prev[n] ~ number)
+                diff[n] = data[n] - data_prev[n]
+            else
+                diff[n] = ""
+            data_prev[n] = data[n]
         }
-        else
-            diff[y] = ""
-        printf(y < nf_max ? OFS : ORS)
+        printf(diff[n])
+        printf(n < nf_max ? OFS : ORS)
     }
 }
+
diff --git a/lin_reg1.awk b/lin_reg1.awk
index ae59c77..c657035 100644
--- a/lin_reg1.awk
+++ b/lin_reg1.awk
@@ -36,7 +36,6 @@ NF > 0 {
             sum_delta[y] += delta1[y]
             sum_delta2[y] += delta0[y]*delta1[y]
 
-
             # x = row, y = col, trendline: y = A + Bx
             for (x=1; x<=nf_max; x++) {
                 count[x,y] += 1
diff --git a/lin_reg2.awk b/lin_reg2.awk
index f06d890..450f0f0 100644
--- a/lin_reg2.awk
+++ b/lin_reg2.awk
@@ -36,7 +36,6 @@ NF > 0 {
             sum_delta[y] += delta1[y]
             sum_delta2[y] += delta0[y]*delta1[y]
 
-
             # x = row, y = col, trendline: y = A + Bx
             for (x=1; x<=nf_max; x++) {
                 count[x,y] += 1
diff --git a/mean.awk b/mean.awk
index 754acfe..069db4c 100644
--- a/mean.awk
+++ b/mean.awk
@@ -5,20 +5,21 @@
 
 BEGIN {
     OFS = FS
-    sign = "[+-]?"
-    decimal = "[0-9]+[.]?[0-9]*"
-    fraction = "[.][0-9]+"
-    exponent = "([Ee]" sign "[0-9]+)?"
-    number = "^" sign "(" decimal "|" fraction ")" exponent "$"
+    #sign = "[+-]?"
+    #decimal = "[0-9]+[.]?[0-9]*"
+    #fraction = "[.][0-9]*"
+    #exponent = "([Ee]" "[+-]?" "[0-9]+)?"
+    number = "^[+-]?([0-9]+[.]?[0-9]*|[.][0-9]*)([Ee][+-]?[0-9]+)?$"
 }
 
 # Welford's 'online' algorithm for variance
-NF > 0 {
+NF {
     for (n=1; n<=NF; n++) {
         if ($n ~ number) {
             count += 1
             (count == 1 || $n < min) ? min = $n : min = min
             (count == 1 || $n > max) ? max = $n : max = max
+            range = max - min
             delta0 = $n - mean
             mean += delta0/count
             delta1 = $n - mean
@@ -29,7 +30,7 @@ NF > 0 {
 }
 
 END {
-    print "mean", "std_err", "std_dev", "min", "max", "total", "count"
-    print mean, sqrt(var/count), sqrt(var), min, max, (mean*count), count
+    print "mean", "std_err", "std_dev", "range", "min", "max", "total", "count"
+    print mean, sqrt(var/count), sqrt(var), range, min, max, (mean*count), count
 }
 
diff --git a/mean_avg.awk b/mean_avg.awk
index e4596b0..bef984d 100644
--- a/mean_avg.awk
+++ b/mean_avg.awk
@@ -20,7 +20,7 @@ NR == 1 {
 }
 
 # Welford's 'online' algorithm for variance
-NF > 0 {
+NF {
     if (NF > max_nf)
         max_nf = NF
     for (n=1; n <= NF; n++) {
@@ -28,6 +28,7 @@ NF > 0 {
             count[n] += 1
             (count[n] == 1 || $n < min[n]) ? min[n] = $n : min[n] = min[n]
             (count[n] == 1 || $n > max[n]) ? max[n] = $n : max[n] = max[n]
+            range[n] = max[n] - min[n]
             delta0[n] = $n - mean[n]
             mean[n] += delta0[n]/count[n]
             delta1[n] = $n - mean[n]
@@ -38,12 +39,12 @@ NF > 0 {
 }
 
 END {
-    print "col", "mean", "std_err", "std_dev", "min", "max", "total", "count"
+    print "col", "mean", "std_err", "std_dev", "range", "min", "max", "total", "count"
     for (n=1; n<=max_nf; n++) {
         if (header[n])
-            print header[n], mean[n], sqrt(var[n]/count[n]), sqrt(var[n]), min[n], max[n], mean[n]*count[n], count[n]
+            print header[n], mean[n], sqrt(var[n]/count[n]), sqrt(var[n]), range[n], min[n], max[n], mean[n]*count[n], count[n]
         else
-            print "col" n, mean[n], sqrt(var[n]/count[n]), sqrt(var[n]), min[n], max[n], mean[n]*count[n], count[n]
+            print "col" n, mean[n], sqrt(var[n]/count[n]), sqrt(var[n]), range[n], min[n], max[n], mean[n]*count[n], count[n]
     }
 }
 
-- 
cgit v1.2.3