So i have created a regression, which regresses a cycle component onto a lagged cycle component and various other variables.

It looks like this: regoffice4 <- dynlm(diff(diff(officecycle,1),1)~lag(diff(diff(subsetoffice$officecycle,1),1),1)+diff(diff(VPIyoy,1),1)+diff(diff(BIP,1),1)+diff(diff(AL-Quote,1),1)+diff(diff(Leerstand Büro,1),1))

when i summarize it with the summary command i get this notification and output:

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.000e+00 4.996e-17 0.000e+00 1.000
lag(diff(diff(subsetoffice$officecycle, 1), 1), 1) 1.000e+00 1.728e-17 5.786e+16 <2e-16 ***
diff(diff(VPIyoy, 1), 1) -3.588e-17 6.723e-17 -5.340e-01 0.596
diff(diff(BIP, 1), 1) -1.227e-18 1.905e-18 -6.440e-01 0.522
diff(diff(AL-Quote, 1), 1) -4.180e-17 1.101e-16 -3.800e-01 0.706
diff(diff(Leerstand Büro, 1), 1) -1.256e-16 1.610e-16 -7.800e-01 0.439

Warning message:
In summary.lm(regoffice4) :
essentially perfect fit: summary may be unreliable

Data looks like the picture posted

-> It would be great if anyone could answer me: have i made a mistake in assembling the regression? or might there be a problem with the data?
I already ran my RSkript before and it yielded me actual results, unlike the problem stated above, where Coefficients looked promising and for example the adj. R^2 wasn´t equal to 1.
So i also thought about maybe i have changed settings? Is that even possible?

Thanks a lot!

I tried to create a dynamic linear model with one lagged variable and expected a logical output. Which i got earlier. However, after restarting it yields different results.