Are thermistors stable in their error?

[NaxFM]

I need a temperature sensor for a thermal chamber and it needs to be cheap, both the sensor and the circuitry.
However the typical 2 degrees of accuracy for a thermistor is just too much, I need something with an accuracy of maximum +-1 Degree C, better if a little less.
I could use a thermistor if I knew the error was stable over time, so for example if the real temperature is 25C and the thermistor always reads 27C it wouldn't be a problem, all i have to do is to calibrate the reading. But it needs to read 27C evey single time the temperature is 25C, even after days.
I can't find this information anywhere, both for thermistor and for integrated temperature sensors, i can easily find the specified accuracy but not the stability.
Do you have any info about this? I just need to know if a thermistor is stable over time.

[nfmax]

Thermistor stability was discussed, along with other interesting topics, in the thread https://www.eevblog.com/forum/metrology/measuring-mk-with-small-rtds/

Thermistors can have excellent temperature stability:

The most stable thermistors are bead thermistors encapsulated in glass. Within the range –20 °C to 60 °C, selected and pre-aged thermistors may be stable to better than a few tenths of a millikelvin per year. Thermistors with resistances in the range 2 kΩ to 10 kΩ also appear to be the most stable. Glass encapsulated disc thermistors and epoxy-encapsulated bead thermistors are also very good but not quite as stable as the glass beads.

This comes from a document published by the BIPM, addressing the suitability of thermistors for metrological applications: https://www.bipm.org/utils/common/pdf/ITS-90/Guide-SecTh-Thermistor-Thermometry.pdf. Another publication I have here reports a test with a 'super stable' epoxy-coated bead thermistor. It showed no detectable drift in resistance/temperature characteristic in the range 0˚C to 60˚C over a 9-month period, at a 1mK uncertainty level: https://www.eevblog.com/forum/metrology/measuring-mk-with-small-rtds/?action=dlattach;attach=1209374. The specific thermistor used in this test was an MEAS 46016, made by TE connectivity: https://www.te.com/usa-en/product-11029033-00.html

I hope this information is of use!

[NaxFM]

Thermistor stability was discussed, along with other interesting topics, in the thread https://www.eevblog.com/forum/metrology/measuring-mk-with-small-rtds/

Thermistors can have excellent temperature stability:

The most stable thermistors are bead thermistors encapsulated in glass. Within the range –20 °C to 60 °C, selected and pre-aged thermistors may be stable to better than a few tenths of a millikelvin per year. Thermistors with resistances in the range 2 kΩ to 10 kΩ also appear to be the most stable. Glass encapsulated disc thermistors and epoxy-encapsulated bead thermistors are also very good but not quite as stable as the glass beads.

This comes from a document published by the BIPM, addressing the suitability of thermistors for metrological applications: https://www.bipm.org/utils/common/pdf/ITS-90/Guide-SecTh-Thermistor-Thermometry.pdf. Another publication I have here reports a test with a 'super stable' epoxy-coated bead thermistor. It showed no detectable drift in resistance/temperature characteristic in the range 0˚C to 60˚C over a 9-month period, at a 1mK uncertainty level: https://www.eevblog.com/forum/metrology/measuring-mk-with-small-rtds/?action=dlattach;attach=1209374. The specific thermistor used in this test was an MEAS 46016, made by TE connectivity: https://www.te.com/usa-en/product-11029033-00.html

I hope this information is of use!

THANK YOU! This is all I needed to know and more!
In the end it seems that even the most common thermistors you can buy for few cents are far more stable that what I need.
Really, I can't thank you enough. I almost had given up on thermistors for my project but this is just perfect!