Dataset source: https://www.worldometers.info/coronavirus/
Data on the server: I have been using couple of tables, this is the main one.
I have created a view to display the values for the last update and the previous day (new cases – previous day new cases) but in the most cases I used DAX to create the measures in the Power BI.
Recovery index measure:
1: Recoveries Total =
CALCULATE (
SUM ( DimCovid[Recoveries] ),
FILTER ( DimCovid, DimCovid[Date] = MAX ( DimCovid[Date] ))
)
2: Cases Total =
CALCULATE (
SUM ( DimCovid[CasesTotal] ),
FILTER ( DimCovid, DimCovid[Date] = MAX ( DimCovid[Date] ))
)
3: RecoveryRate =
DIVIDE ( [Recoveries Total], [Cases Total], 0)
Let’s see how these measures got created.
7-day median/100K
1:
Population = CALCULATE ( SUM ( WorldometersOnline[Population] ) )
2:
PopRate = [Population]/100000
3:
7-day median/100K =
VARmed =
CONVERT (
CALCULATE (
MEDIAN ( VwDimCovid[NewCases] ),
DATESINPERIOD ( VwDimCovid[Date], LASTDATE ( VwDimCovid[Date] ), -7, DAY)
),
INTEGER
)
RETURN
DIVIDE (med, [PopRate], 0)
Daily limit/100K
Safe case threshold according the EU regulation is <25/14 days. Per 1 day is 25/14=1.786
Limit =
VAR lim =
DIVIDE ( [7-day median/100K], 1.786, 0)
RETURN
IF (
lim > 1,
FORMAT (lim, "0.00") & "x over limit",
IF (
lim = 1,
FORMAT (lim, "0.00") & "x =limit",
FORMAT (lim, "0.00") & "x under limit"
)
)
RN- reproductive number
New case over previous cases during some time period (7 days)
1: New Cases Last 7 Days =
CALCULATE (
SUM ( 'VWDimCovid'[NewCases] ),
DATESINPERIOD ( 'VWDimCovid'[Date], MAX ( vWDimCovid[date] ), -7, DAY )
)
2:
New Cases 8th-14th Day =
CALCULATE (
SUM ( 'VWDimCovid'[NewCases] ),
DATESINPERIOD ( 'VWDimCovid'[Date], LASTDATE ( vWDimCovid[date] ) - 7, -7, DAY)
)
3:
RN = DIVIDE([New Cases Last 7 Days],[New Cases 8th-14th Day],0)
Progress to zero
Corresponds to the percentage decline from a previously recorded peak level. The metric ranges from 0% (representing a geography that has not yet peaked) to 100% (representing a geography wherein 0 cases have been recorded for at least seven days)
1:
World7-day/avg Newcase =
CALCULATE (
AVERAGE ( vWDimCovid[NewCases] ),
DATESINPERIOD ( vWDimCovid[date], LASTDATE ( vWDimCovid[date] ), -7, DAY)
)
2:
AverageIncedenceMAXX =
MAXX (
SUMMARIZE (
vWDimCovid,
vWDimCovid[date],
"XNC", AVERAGE ( vWDimCovid[NewCases] )
),
[World7-day/avg Newcase]
)
3:
Progress to zero =
1 - ( DIVIDE ( [World7-day/avg Newcase], [AverageIncedenceMAXX], 0 ) )
progress to zero in the tooltip
Spread frequency
I was a little bit lazy to figure out average for my dataset, so I did it old statistic way…
1: InfectionIncidence =
VARday1 =
CALCULATE (
SUM ( VWDimCovid[NewCases] ),
FILTER ( vWDimCovid, VWDimCovid[date] = MAX ( VWDimCovid[date] ))
)
VARday2 =
CALCULATE (
SUM ( VWDimCovid[NewCases] ),
FILTER ( vWDimCovid, VWDimCovid[date] = MAX ( VWDimCovid[date] ) - 1)
)
VARday3 =
CALCULATE (
SUM ( VWDimCovid[NewCases] ),
FILTER ( vWDimCovid, VWDimCovid[date] = MAX ( VWDimCovid[date] ) - 2)
)
VARday4 =
CALCULATE (
SUM ( VWDimCovid[NewCases] ),
FILTER ( vWDimCovid, VWDimCovid[date] = MAX ( VWDimCovid[date] ) - 3)
)
VARday5 =
CALCULATE (
SUM ( VWDimCovid[NewCases] ),
FILTER ( vWDimCovid, VWDimCovid[date] = MAX ( VWDimCovid[date] ) - 4)
)
VARday6 =
CALCULATE (
SUM ( VWDimCovid[NewCases] ),
FILTER ( vWDimCovid, VWDimCovid[date] = MAX ( VWDimCovid[date] ) - 5)
)
VARday7 =
CALCULATE (
SUM ( VWDimCovid[NewCases] ),
FILTER ( vWDimCovid, VWDimCovid[date] = MAX ( VWDimCovid[date] ) - 6)
)
RETURN
(day1 + day2 + day3 + day4 + day5 + day6 + day7) / 7
{86400 = number of seconds in 1 day}
2:
String Duration Incedence Trend in Days Hours Minutes and Seconds =
VARvSeconds =
INT (DIVIDE (86400, [InfectionIncidence] ))
VARvMinutes =
INT (vSeconds / 60)
VARvRemainingSeconds =
MOD (vSeconds, 60)
VARvHours =
INT (vMinutes / 60)
VARvRemainingMinutes =
MOD (vMinutes, 60)
VARvDays =
INT (vHours / 24)
VARvRemainingHours =
MOD (vHours, 24)
RETURN
vDays & " Days , " & vRemainingHours & " Hrs : " & vRemainingMinutes & " Min : " & vRemainingSeconds & " Sec"
measure in the tooltip
Region / District: New cases in the last 14 days/100K
If no district is filtered the chart displays data for the region, if district is filtered the chart will show the data for a specific district.
1:
2-Week/NewCases =
CALCULATE (
SUM ( DistrictGeoData[NewCases] ),
DATESINPERIOD ( DistrictGeoData[Date], MAX ( DistrictGeoData[Date] ), -14, DAY)
)
2:
NI14-day Incid/100K =
IF (
HASONEVALUE ( DistrictGeoData[District] ),
CALCULATE (DIVIDE ( [2-Week/NewCases], DistrictGeoData[DistrictPopRate] )),
CALCULATE (
DIVIDE ( [2-Week/NewCases], DistrictGeoData[RegionPopRate] ),
DistrictGeoData[Region] = "Nitriansky"
)
)
Change (KPI in the red frame) Indicates the change from the previous day.
You need a dataset with consecutive dates otherwise this code won't be working correctly.
1:
Daily Confirmed =
CALCULATE (
SUM ( SlovakiaDetails[DailyConfirmed] ),
FILTER (
SlovakiaDetails,
SlovakiaDetails[Date] = MAX ( SlovakiaDetails[Date] )
)
)
2:
Daily Confirmed day before =
CALCULATE (
SUM ( SlovakiaDetails[DailyConfirmed] ),
FILTER (
SlovakiaDetails,
SlovakiaDetails[Date]
= MAX ( SlovakiaDetails[Date] ) - 1
)
)
3:
Daily confirmed change =
[Daily Confirmed] - [Daily Confirmed day before]
4:
DailyConfComment =
VAR Down =
UNICHAR (10134)
VAR Up =
UNICHAR (119108)
VAR Side =
UNICHAR (129094)
RETURN
IF (
[Daily confirmed change] > 0,
"(" & FORMAT ( [Daily confirmed change], "+#,##0") & ")",
IF (
[Daily confirmed change] < 0,
"(" & FORMAT ( [Daily confirmed change], "#,##0") & ")",
Side & " "
& FORMAT ( [Daily confirmed change], "#,##0")
)
)
Slovakia herd immunity
I used 2 open sources data from GitHub (.csv) to create this visual.
I took dates from the calendar table
calculating daily new cases from PCR and AG table
1:
PCRDennePrirastky =
CALCULATE (
SUM ( PositiveTests_District[DennychPrirastkov] ),
FILTER ( 'Calendar', 'Calendar'[Date] = MAX ( PositiveTests_District[Datum] ))
)
2:
AG = CALCULATE(SUM(AgTests_District[Ag_Pos]))
I attached the measures to the table
Calculating running totals for PCR and AG test results. I created a new column for AG tests (1.8 * AG) because AG tests can only detect 50-55% of what PCR tests would detect in a similar circumstances. Don't worry about this statement.
3:
Ag_spolu =
CALCULATE (
SUM ( [50%AGplus] ),
FILTER ( Promorovanost, EARLIER ( Promorovanost[Date] ) >= Promorovanost[Date] )
)
4:
PCR_spolu =
CALCULATE (
SUM ( [PCR] ),
FILTER ( Promorovanost, EARLIER ( Promorovanost[Date] ) >= Promorovanost[Date] )
)
5: (sum of values from PCR and AG running totals )
AG_PCR_Spolu = [AG_spolu]+[PCR_spolu]
6: (Calculating the herd imunity)
%_Premorovanost =
DIVIDE ( [AG_PCR_Spolu], [SR_population], 0)
Newly created table for the visual
Comentarios