Electroencephalogram (EEG) signal has been widely used to analyze brain activities so as to diagnose certain brain-related diseases. They are usually recorded for a fairly long interval with adequate resolution, which requires considerable amount of memory space for storage and transmission. Compression techniques are necessary to reduce the signal size. As compared to lossless compression techniques, lossy compression techniques would provide much higher compression ratio (CR) by taking advantage of the limitation of human perception. However, that is achieved at the cost of introducing more compression distortion, which reduces the fidelity of EEG signals. How to select a suitable lossy EEG compression technique? This motivates us to survey those existing lossy compression algorithms reported in the last two decades. We attempt to analyze the algorithms and provide a qualitative comparison among them.