I made my ACS one-year samples from 2007 through 2018 into county level data using statefip and countyfip. Then I see 144 out of 523 FIPS are missing complete set of observations for some consecutive years. 99 FIPS do not have 2007-2011 data (for example FIPS 05 125), and 45 FIPS do not have 2012-2018 data (for example 01 125). However, it looks like the data for those FIPS exists when I search the Census data page.
Could there be any way to incorporate those missing data?
As noted on the description tab for COUNTYFIP, counties are not identified in the public use microdata samples (PUMS) beginning in 1950. IPUMS USA identifies counties, where possible, from other low-level geographic identifiers. If you look at the list of identified counties linked at the bottom of the COUNTYFIP description (a downloadable Excel file), you will see that Saline County (COUNTYFIP 125) in Arkansas (STATEFIP 05) is not available in 2006-2011 and that Tuscaloose County (COUNTYFIP 125) in Alabama (STATEFIP 01) is not available in 2012-forward.
If you are interested in data aggregated at the county rather than person-level microdata, I suggest looking at IPUMS NHGIS to see if there is a summary table that is appropriate.