Would you please review a few things below? This problem lends itself well to using helper cells, as certain values need to be more readily accessible. Even with helper cells, the formulas become rather messy. I've manually assigned a row index (idx) to the original baseline scoring table, beginning with the most negative score zone and ending with the most positive one. Then using formulas, the scoring table is recreated in a more useful format, so that the starting score (strt) and the limit (lim) for each scoring zone are readily accessible. Another column ("Cond Met?) in the helper table performs logic checks to determine whether the "Status" shown in F23:G23 (e.g., >50 and Slowing) are found in the original scoring table descriptions, and if so, where. Since the last two rows of the scoring table have one additional logic check, the formulas include a feature to determine whether the 3rd or 4th condition should be applied, based on the row number.
I noticed several inconsistencies that lead me to question whether the offering below will really do what you want it to do.
1. The scores for the last two rows of the original scoring table have been swapped...could you confirm that doing so is correct? It seemed to me that >50 is associated with positive scoring zones elsewhere, and <50 is associated with negative scoring zones elsewhere, so I swapped the two scores.
2. Doing so (swapping rows 5 and 6) resolved some inconsistencies in the example table...please review the "Score Bounds" in that table and reconcile those with the text in the "Note" column. There are places where -8 and +8 are mentioned that appear to be incorrect. And the "Change" column shows 3 and 1 in 5th and 6th rows, but those are not consistent with the corresponding notes.
3. Please confirm what the additional criteria are for the 5th and 6th rows in the scoring table. I believe there are conflicting definitions shown when comparing cells E6:E7 to P15:P16: "for 1st time" vs. ">=big(k=3)"? Which is it? On this last item, does "big" mean ">=LARGE(B2:B1500,3)"?...you want to compare the Last Value to the 3rd largest value in column B? Or do you want to know whether the "Peak >50 and Slowing" has occurred for the first time in Column B?
I copied parts of the example table immediately below the original one and swapped elements in the 5th and 6th rows (score bounds and change). Please confirm that these swaps are what was intended based on your notes in the original table. The pink cells in O18:O23 use a formula to calculate a score for the conditions shown in the revamped example table. If these all check out okay, then let me know and I'll clean up the sheet so that the F10 scoring formula performs the same calculation, but pointing to the correct inputs elsewhere on the sheet.
Other observations:
1. None of the formula that you've used in the sheet need to be array formulas. I've eliminated the array formulas in this example.
2. The use of merged cells is highly discouraged. They create several unintended problems (e.g., when attempting to click on ranges for formula development, the desired range cannot be selected). I've eliminated several merged cells...and in nearly all cases, a simple multi-cell selection and then reformat to "center across selection" creates a heading that effectively spans the same cells.
formula problem_rev.xlsx |
---|
|
---|
| A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W |
---|
1 | DATE | PMI | Change | | PMI | Score | idx | Cond Met? | | idx | >= | Start and < | | | | | | | | | |
---|
2 | Sep-92 | 50.44 | | | >50 and growing | | | +5 to +8 | 5 | FALSE | | 1 | -10 | -8 | | | | | | | | | |
---|
3 | Dec-92 | 47.59 | -2.85 | | >50 and Slowing | | | +5 to 0 | 4 | TRUE | | 2 | -8 | -5 | | >50 and growing | >50 and slowing | <50 and slowing | <50 and growing | Peak >50,slowing for 1st time | Trough <50,growing for 1st time | | |
---|
4 | Mar-93 | 52.15 | 4.56 | | <50 and slowing | | | -5 to -8 | 2 | FALSE | | 3 | 0 | -5 | | | |
---|
5 | Jun-93 | 55.36 | 3.21 | | <50 and growing | | | -5 to 0 | 3 | FALSE | | 4 | 0 | 5 | | | |
---|
6 | Sep-93 | 57.69 | 2.33 | | Peak >50,slowing for 1st time | | | +8 to +10 | 6 | FALSE | S | 5 | 8 | 5 | | +5 to +8 | +5 to 0 | -5 to -8 | -5 to 0 | -8 to -10 | +8 to +10 | | |
---|
7 | Dec-93 | 59.47 | 1.78 | | Trough <50,growing for 1st time | | | -8 to -10 | 1 | FALSE | S | 6 | 10 | 8 | | | | | | | | | |
---|
8 | Mar-94 | 61.77 | 2.30 | | | | | | 4 | <-- use idx | | | | | | | | | | | | | |
---|
9 | Jun-94 | 61.99 | 0.22 | | | | | -8 to -10 | | | | | | | | Examples |
---|
10 | Sep-94 | 61.49 | -0.50 | | FINAL SCORE | 1 | | | | | | | | | | Condition | Status | Score Bounds | PMI | Change | Score | Note |
---|
11 | Dec-94 | 59.81 | -1.68 | | | | | Percentile | | | | | | | | | >50 | Slowing | +5 to 0 | 56 | -2.6 | 2 | Because it's +5 (-2,6 that is the change),so the score is +2 |
---|
12 | Mar-95 | 54.3 | -5.51 | | Last PMI | 52.1 | | 58% | | | | | | | | | >50 | Growing | +5 to +8 | 54 | 5.1 | 8 | Because it's +5 (+5,1 that is the change) but as the maximum score allowed is +8,the score is +8 |
---|
13 | Jun-95 | 50.72 | -3.58 | | | | | | | | | | | | | | <50 | Slowing | -5 to -8 | 43 | -4.2 | -8 | Because it's -5(-4,2 that is the change),but as the maximum score allowed is -8,the score is -8(and not -9 ) |
---|
14 | Sep-95 | 48.44 | -2.28 | | | | | | | | | | | | | | <50 | Growing | -5 to 0 | 45 | 2.4 | -3 | Because it's -5(+2,4),so the score is -3 |
---|
15 | Dec-95 | 43.88 | -4.56 | | ManufacturingPMI | | | | | | | | | | | Last Value>= big(k=3) of Row B | >50 | Slowing | -8 to -10 | 63 | 3 | 9 | Because it's +8(+1 that is the change),he score is +9 |
---|
16 | Mar-96 | 45.19 | 1.31 | | Last - 1 | 56.3 | MIN | 30.86 | | | | | | | | Last Value <= little(K=3) of row B | <50 | Growing | +8 to +10 | 32 | 1 | -10 | Because it's -8(-3 that is the change),but as the max score allowed is -10 ,the score is -10(not -11) |
---|
17 | Jun-96 | 46.34 | 1.15 | | Last Value | 52.1 | Average | 51 | | | | | | | | | start | equiv I8 | | | equiv f18 | | |
---|
18 | Sep-96 | 47.27 | 0.93 | | Change | -4.2 | MAX | 63.1 | | | | | | 0 | 2 | 2 | 5 | 4 | +5 to 0 | | -2.6 | | |
---|
19 | Dec-96 | 49.36 | 2.09 | | Large(B:B,3) | 61.99 | FALSE | delete later | | | | | | 8 | 8 | 10 | 5 | 5 | +5 to +8 | | 5.1 | | |
---|
20 | Mar-97 | 51.11 | 1.75 | | Small(B:B,3) | 32.02 | FALSE | delete later | | | | | | -8 | -8 | -10 | -5 | 2 | -5 to -8 | | -4.2 | | |
---|
21 | Jun-97 | 51.49 | 0.38 | | | | | | | | | | | 0 | -3 | -3 | -5 | 3 | -5 to 0 | | 2.4 | | |
---|
22 | Sep-97 | 50.63 | -0.86 | | | Status | | | | | | | | 10 | 9 | 9 | 8 | 6 | +8 to +10 | | 1 | | |
---|
23 | Dec-97 | 51.06 | 0.43 | | | >50 | Slowing | | | | | | | -10 | -10 | -11 | -8 | 1 | -8 to -10 | | -3 | | |
---|
24 | Mar-98 | 50.38 | -0.68 | | | | | | | | | | | | | | | | | | | | |
---|
|
---|