DAYS360 in WPS Spreadsheet: Compare 30/360 Methods with Actual Calendar Days
Calculate WPS 30/360 day counts using explicit US or European methods, test a month-end difference, and keep conventional units separate from actual elapsed days.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer: DAYS360(start_date, end_date, [method]) measures an interval on a twelve-month, thirty-days-per-month convention. FALSE or omission selects the documented US/NASD method; TRUE selects the European method. It is not a count of actual calendar days.
Availability and version history
- Introduced: documented by 2019-05-23 in the official WPS lesson; original introduction is not established.
- Availability: WPS Office for Windows Personal; the lesson metadata gives no numeric release.
- Scope: exact current supported builds remain unverified; the historical Windows lesson does not establish all WPS clients.
- Checked: October 14, 2026 against complete function-specific documentation. Examples are independent expectations; no native WPS test was performed.
Syntax and method selection
=DAYS360(start_date, end_date, [method])
The two dates are required. The WPS walkthrough explicitly permits omission of the method, although its compact syntax shows all three argument names. Square brackets describe that optional argument and are not typed.
Use actual date values or DATE expressions. Results below are independent models of the documented conventions, not native WPS observations. Keep the output numeric and label it as a 30/360 count. This guide demonstrates function behavior; choosing a contractual day-count convention is a separate decision.
Example 1: compare a conventional month with real days
=DAYS360(DATE(2026,1,15),DATE(2026,2,15),FALSE)
=DAYS360(DATE(2026,1,15),DATE(2026,2,15),TRUE)
Expected results are 30 and 30. The actual calendar interval is 31 days, which can be checked by subtracting the two DATE values.
The corresponding 15 February to 15 March 2026 interval also gives 30 under either convention, while the actual interval is 28 days. Do not call either conventional result a Gregorian elapsed-day count.
Example 2: calculate an ordinary multi-month interval
=DAYS360(DATE(2025,12,15),DATE(2026,2,20),TRUE)
Expected result: 65, calculated as one year×360 plus (2−12)×30 plus (20−15). These ordinary dates avoid endpoint-adjustment ambiguity. The actual calendar difference is 67 days.
Reversing these ordinary endpoints gives −65 under the convention. The source explicitly states that a start later than the end produces a negative count. Identical ordinary endpoints give zero.
Advanced example: see a real difference between methods
The official embedded method table says the European method changes a day numbered 31 to day 30. For the US method, it describes a month-end start as day 30 and a month-end end date with an earlier-than-30 start as the first of the following month.
Use a start of 29 Jan 2026 and an end of 31 Jan 2026:
=DAYS360(DATE(2026,1,29),DATE(2026,1,31),FALSE)
=DAYS360(DATE(2026,1,29),DATE(2026,1,31),TRUE)
Documented-model expectations are 2 and 1. The US case moves the end to 1 February; the European case changes it to 30 January. Omitting the method follows the documented FALSE default, so its corresponding expectation is 2.
This is why the method must be preserved when copying a formula. Shared input dates do not imply the same conventional result.
Advanced example: build a labeled comparison table
Enter the following date-only pairs using DATE and compare both methods alongside ordinary end-minus-start arithmetic:
| Start | End | US/NASD | European | Actual days |
|---|---|---|---|---|
| 15 Jan 2026 | 15 Feb 2026 | 30 | 30 | 31 |
| 15 Feb 2026 | 15 Mar 2026 | 30 | 30 | 28 |
| 29 Jan 2026 | 31 Jan 2026 | 2 | 1 | 2 |
| 31 Jan 2026 | 31 Mar 2026 | 60 | 60 | 59 |
The last row uses documented month-end/day-31 adjustments. A conventional year from 15 Jan 2026 to 15 Jan 2027 gives 360, while the actual interval is 365. Keep the count’s convention attached to any exported number.
Ambiguities and errors left open
The Chinese US-method table’s “otherwise” clause is compressed and could be read too broadly. This guide limits month-end examples to the explicit branches above, rather than claiming a complete certified implementation for every February, reversed endpoint or time-fraction case.
Unspecified method coercion, invalid-date error codes, fractional times, imported date systems and every February edge remain unverified. The repeated historical serial-origin explanation also has inconsistent day-count wording, so hard-coded serial-base examples are avoided.
Use DAYS for actual numeric date differences and DATE for reproducible components. A 360-day convention should never be silently substituted for actual elapsed time.
References and verification
- WPS Academy: DAYS360, 2019-05-23 (Chinese; Windows Personal).
- WPS Academy: DATE.
- Editorial testing standards.
The complete function definition and walkthrough were reviewed. Worked outputs are independent calendar and time models; source ambiguities and native boundaries remain explicit.